ArticlePDF Available

Institutions and intellectual property: The influence of institutional forces on university patenting


Abstract and Figures

Over the past 20 years, the number of patents assigned to universities has increased dramatically. This increase coincided with several policy initiatives, such as the Bayh-Dole Act of 1980, designed to foster technology transfer between universities and the private sector. This paper examines the effect of such policies using an institutional framework, designed to illustrate how factors both from inside and outside of academia influence the decision to patent university research. We find passage of the Bayh-Dole Act spurred university patenting, but did not induce additional applied research funding. Thus, Bayh-Dole fostered technology transfer, but did not result in more applied research at universities. © 2005 by the Association for Public Policy Analysis and Management
Content may be subject to copyright.
Yixin Dai
David Popp
Stuart Bretschneider
Institutions and Intellectual
Property: The Influence of
Institutional Forces on
University Patenting
Journal of Policy Analysis and Management, Vol. 24, No. 3, 579–598 (2005)
© 2005 by the Association for Public Policy Analysis and Management
Published by Wiley Periodicals, Inc. Published online in Wiley InterScience (
DOI: 10.1002/pam.20116
Manuscript received October 2004; review complete November 2004; revision complete November 2004; accepted Jan-
uary 2005.
Over the past 20 years, the number of patents assigned to universities has increased
dramatically. This increase coincided with several policy initiatives, such as the
Bayh-Dole Act of 1980, designed to foster technology transfer between universities
and the private sector. This paper examines the effect of such policies using an
institutional framework, designed to illustrate how factors both from inside and
outside of academia influence the decision to patent university research. We find
passage of the Bayh-Dole Act spurred university patenting, but did not induce addi-
tional applied research funding. Thus, Bayh-Dole fostered technology transfer, but
did not result in more applied research at universities. © 2005 by the Association
for Public Policy Analysis and Management
University and industry research are fundamentally different. In most cases, finan-
cial support for university research comes from the government or not-for-profit
organizations. University research is driven by individual academic research inter-
ests, while industry research is driven by potential market benefits and corporate
decision processes. The usual output of university research is knowledge that is typ-
ically diffused through scientific publications. In contrast, industrial research and
development (R&D) activities focus on technology applications, which mark com-
mercially usable technologies as their outcome.
Over time, however, outcomes from university research have changed. In addition
to scientific papers, commercial outputs like patents, trademarks, and other intel-
lectual properties have become more common. The Bayh-Dole Patent and Trade-
mark Amendments Act of 1980 made it easier for universities to patent the results
of federally funded research. Prior to passage of the Act, federally funded research
results could only be patented with government permission. Receiving such per-
mission was complicated, as policies for granting permission varied by agency. For
example, both the National Science Foundation (NSF) and Department of Health,
Education, and Welfare (now Health and Human Services, which contains the
National Institutes of Health) only granted waivers on a case-by-case basis, so that
universities would need to negotiate with the agency each time they wanted to
patent sponsored research (Mowrey et al., 2001). The Bayh-Dole Act created a uni-
580 / The Influence of Institutional Forces on University Patenting
fied policy, allowing the results of federally funded research to be patented without
the need for receiving a specific waiver. By both simplifying the path to patenting
federally funded research and providing a signal of Congress’ support for the impor-
tance of patenting such research, the number of university-issued patents grew dra-
matically after passage of Bayh-Dole, as shown in Figure 1. During this time period,
university patenting grew at a rate nearly ten times higher than patents from indus-
trial sources. At the same time, universities increased their patenting per R&D dol-
lar during a period in which overall patenting per R&D dollar was declining (Mow-
ery et al., 2001).
While the final decision about research done in universities remains centered on
the individual researcher, recent studies show these decisions are influenced by the
university environment, societal forces, and policy. First, policies such as Bayh-Dole
and the National Cooperative Research Act of 1984 are designed to encourage tech-
nology transfer between universities and industry. Second, universities are closely
connected with societal demands to promote economic development. Numerous
studies focus on the role university research plays in industrial innovation and
national economic development (Salter & Martin, 2001; Audretsch, Link, & Scott,
2002). Third, universities themselves are giving additional attention to potential
economic benefits inherited from their research, even in basic research (National
Science Foundation [NSF], 2002b; Jaffe & Trajtenberg, 1996).
Indeed, these forces through which university researchers are encouraged to
increase patenting are linked. During the 1980s, concerns over international com-
petitiveness led to several policy initiatives to increase the commercial relevance of
basic research institutions by encouraging technology transfer to industry (Boze-
man, 2000). As a result, in this paper we present an integrated framework for
examining how societal forces push both university and governmental policies
toward using university patents as a valid research outcome. Using the perspective
of institutional change and learning, our framework suggests that societal forces
Figure 1. Number of university-issued patents (1975–1997).
The Influence of Institutional Forces on University Patenting / 581
create incentives for universities to adjust their structures and incentives through
policy changes influencing the research decisions of faculty. These changes influ-
ence not only the focus of faculty research, but also the form of the resulting
research output.
Beginning with a focus on the behavior of individual researchers, we present a
model illustrating how various stages in the research process affect the decision on
how to diffuse results (for example, publications, patents). Crucial to this decision
is the nature of the research project. The choice of research project will be shaped
by social forces such as alternatives available for research funding. For example,
concerns over economic competitiveness in the 1980s led to an increase in applied
research funding to universities. Thus, we next analyze institutional factors outside
the research process—such as policy, research funding decisions, and university
culture—that influence both the decision on what type of research to do and the
choice of a diffusion method. Many things affect university community culture and
influence an academic researcher’s decision to patent, such as an academic depart-
ment director’s attitude toward patenting (Thursby, Jensen, & Thursby, 2001) and
efforts by Technology Transfer Offices to encourage patenting by faculty members
(Mowery et al., 2001; Hall, Jaffe, & Trajtenberg, 2001). We combine the individual
level and institutional/societal level models into a single theoretical framework
explaining both research funding decisions and university patenting. Using data
from 1970 to 1997, we test this framework empirically, examining both the forces
that change university patenting behavior and changes in the sources of university
funding. We conclude with implications for both policy and for future research.
Recent research suggests that universities have become more focused on market
forces and have responded by developing an “entrepreneurial university,” contain-
ing research groups that have firm-like qualities. Etzkowitz (2003) attributes these
changes to competitive funding distribution and the “inner logic” universities devel-
oped during the process of expanding the focus of academia from teaching to
research. As a result, university research faces multiple influences from outside aca-
demia, affecting not just the decision of whether to publish or patent research
results, but the whole research process. Since the decision of a diffusion method
reflects these influences on the research process, in this section we develop a model
of the whole research process—from the initial generation of an idea to dissemina-
tion of the results.
Figure 2 presents a synthesis of both the individual and institutional levels of
influence on research decisionmaking. The boxed portion of the figure represents
the research production of an individual researcher. As noted in the figure, this
process takes place in a broader societal context, represented by concentric circles
of influence. In the sections that follow, both the research production process and
these concentric circles of influence are discussed.
University researchers usually process multiple research activities in parallel, so
that they are often doing research and looking for funding at the same time.
Nonetheless, for each research project, the research process follows a stepwise pro-
cedure: knowledge exploration and diffusion of results. The boxed portion of Figure
2 provides a schematic of this process. The knowledge exploration process starts
with the generation of a research idea. Next comes the targeting and securing of
financial support, such as research grants. Typically, the nature of the research idea
determines the possible funding resources. But in some cases, researchers need to
582 / The Influence of Institutional Forces on University Patenting
tailor their research ideas to suit special funding requirements. On the other hand,
researchers can alter funding allocations by using their influence in the science
community—that is, outstanding researchers have demonstrated significant ability
in influencing research foci in society.1After getting financial support, researchers
devote their time and energy to knowledge generation and obtaining final results.
Diffusion, in which the results of research are made “visible” to the scientific com-
munity, is the last phase of the research process. Through diffusion, the results gain
recognition both at the university and societal levels. Arrows in the figure reflect the
influence flows within the research process.
The research diffusion decision, which refers to the decision on what form (or
forms) the researcher chooses to present a research result, is far from a single
choice process. Possible diffusion methods include publications, lectures, confer-
ence papers, patents, commercial secrets, reports, and demonstration projects.
Researchers may choose several methods simultaneously. The goals of diffusion
include maintaining freedom of future action, raising one’s scientific reputation,
gaining future funding, or leveraging knowledge to the society (Owen-Smith & Pow-
ell, 2001). Among these, securing future financial support is of high importance.
Researchers want to convince other scientists, the science community, or policy
makers that their research results are valuable and thus deserving of future finan-
cial support (Moore, 1996).2
Figure 2. Societal influences and the research process.
1For example, NSF and other funding organizations have researchers meet to generate agendas for pro-
grams or invite researchers to peer review proposals. In these ways, particular researchers can channel
funding allocation according to their individual research interests.
2 As evidence of the importance of funding, studies show a strong relationship between the number of
publications and securing funding for research (Harris, 1984). Furthermore, Wodarski (1990) notes that
researchers with higher prestige are more likely to get financial support.
The Influence of Institutional Forces on University Patenting / 583
Influence Flows in the Individual Research Process
As shown in Figure 2, decisions at each stage of the individual research process may
affect the diffusion decision. First, the choice of a research idea could influence the
diffusion method directly (path 1). For example, the outcome of basic research is
typically a publication or conference paper, while applied research ideas may be dif-
fused via patents, trademarks, or other forms of commercial usage. Second,
research funding may directly influence the choice of diffusion method (path 2).
For example, public research funding requires a research report or publishable
paper as a final product. The third flow (path 3) represents the sequence of deci-
sions made throughout a research project. In every step, the researcher may face
alternative choices about the diffusion method, and thus change their diffusion
decisions. This is particularly true as more research projects combine basic and
applied research elements. For instance, research on computer software contains
both a basic research element (that is, improving an algorithm) and an applied ele-
ment (that is, commercial value). Thus, both basic funding and applied funding
may help a researcher finish the project. Such a research project could result in
both publications and patents. Last, availability of funding could influence the gen-
eration of a research idea and in turn influence the diffusion method (path 4).
To understand how these flows influence the diffusion decision, we consider the
role of different types of research. Research is often classified as “basic” or
“applied.” When the distinction is clear-cut, the influence flow is simple and
straightforward. However, as more and more research fits into both basic and
applied categories, multiple diffusion methods become more likely.
Basic Research Processes
Basic research, which is usually initiated by individual research interest, focuses on
pure theoretical knowledge rather than commercial benefit. The generation of basic
research ideas is individually centered. University researchers are motivated by a
“sacred spark” (Cole & Cole, 1973). Basic research results are typically not directly
appropriate for commercial applications. Consequently, the choice of diffusion
method flows directly from the selection of the research idea or as a direct conse-
quence of the process, with little or no regard to research funding (path 1 in Figure 2).
Because of its public goods nature, most basic research is funded by the federal gov-
ernment and performed at universities or federally funded research and development
centers. Since World War II, basic research funding, especially for university research,
comes mostly from federal government initiatives and funding (Shapley & Roy, 1985).
Universities play a major role in doing basic research. Forty-three percent of the total
basic research in the United States is performed in the academic sector. In 2000, the
federal government accounted for 58% of total academic R&D funding, among which
74% went to basic research (National Science Foundation, 2002b). Figure 3 shows the
share of federal government funding to the academic sector since 1954.
Similarly, basic research comprises the majority of research performed by uni-
versities. Influenced by Vannevar Bush’s report “Science—the Endless Frontier,”
university research has generally followed the rule “basic-research-is-best” (Shapley
& Roy, 1985). Since 1955, basic research has consistently comprised over 50% of
total academic research and development expenditures. The share of basic research
peaked at 76.9% in 1968 and 1969. Despite the increased importance in applied
research at universities, the share of basic research was still at 68.5% in 2000
(National Science Foundation, 2002b).
584 / The Influence of Institutional Forces on University Patenting
Applied Research Processes
While basic research continues to dominate academic work, applied research has
grown in importance to universities since the early 1970s. The NSF defines applied
research as research aimed at gaining knowledge or understanding to determine the
means by which a specific, recognized need may be met (National Science Foun-
dation, 1996).3Such research is often designed to generate future market value.
However, this need not be the case. While large research programs in space and
defense lead to products designed to meet a specific, recognized need, limited pri-
vate markets for these products may lower future market value. Such applied
research projects are thus less likely to be performed by industry alone. In either
case, the dissemination process for applied research results should provide outputs
leading to more widespread usage of these results, such as patents, trademarks,
industry reports, or demonstration projects.
However, the dissemination decision is often more complex for academic
researchers. Decisions on patenting do not happen naturally in universities. As
with basic research, many academics performing applied research still use publi-
cations as a main output due to academic inertia. Even if the researcher decides to
patent, it is unlikely that this decision was made at the very beginning of the
research process.
Nonetheless, universities do patent, and the number of university patents has
been growing over time. The number of successful university patent applications
increased from 517 in 1980 to 3,289 by 1995. These patents enable universities to
enter licensing agreements with industrial partners that can develop commercially
viable products from university research. Indeed, a primary motivation of the Bayh-
Dole Act was to encourage such technology transfer, so that the results of university
Figure 3. Share of university research funding (1964–2000).
3 Although the NSF provides official definitions for basic research, applied research and development in
its 1996 Science and Engineering Indicators and has applied these definitions throughout its data col-
lection, we are aware that discussions about the appropriateness of those research classifications con-
tinue. Link (1996) discusses these research classifications for industrial R&D and indicates that the
majority of the firms agree with NSF’s definition. However, in larger firms, disagreements were reported
with regards to the definition of development.
The Influence of Institutional Forces on University Patenting / 585
research could have greater commercial impact. These licensing efforts have
increased along with university patenting. Licensing revenues to Association of
University Technology Managers member universities increased from $222 million
in fiscal year 1991 to $698 million in fiscal year 1997 (Association of University
Technology Managers, 1998, cited in Mowery et al., 2001).
Mixed Research Processes
Recently, more and more research activity has included both basic and applied
characteristics. This is especially true for newly emergent research areas in inter-
disciplinary fields. For example, human genome research both extended common
knowledge about the identification and sequencing of human DNA, which was a
theoretical milestone, and led to new medical applications, often as the result of
technology transfer to the private sector.4
In a mixed research process, the dissemination decision is not solely decided by
the research idea or the source of funding. Many internal and external factors influ-
ence the decision through multiple channels. Both the motivations and results
derived from mixed research may be basic or applied. For example, research to
design new computer software can be inspired by the development of calculation
theory in mathematics or by the demand to solve some real problem in electronic
engineering. As mixed research contains elements of both basic and applied
research, researchers may seek financial support from either basic or applied
research funding sources. Diffusion options for mixed research are generally
broader and likely to include multiple methods of diffusion. As such, mixed research
processes are likely to include any or all of the four paths presented in Figure 2.
The previous section describes links between the research process and the choice of
diffusion method. However, institutional factors outside the research process, such
as university culture and technology policy of federal or state government, also
influence this decision. The concentric circles of Figure 2 present a three-level
model, representing society, university, and individual levels of influence. Each cir-
cle in the diagram represents institutional boundaries in the real world. In this
model, individual researchers are embedded in a university research culture. Many
values that affect a researcher’s actions derive directly from this environment. In the
same way, universities are embedded in society whose norms and rules influence
universities and in turn, individuals within universities.
Societal Influences on Individual Decisions
Our concept of society includes scientific, economic, and political communities and
their interactions. Changing regimes in economic or political communities change
research agendas and provide “hot spots” in the science community. For instance,
increasing energy prices during the energy crisis in the early 1970s led to dramatic
increases in the level of energy R&D (Popp, 2002). Such societal changes influence
both the idea generation and funding stages of the research process, as well as the
4 Information from Department of Energy’s Genomics: GTL information Web site:
586 / The Influence of Institutional Forces on University Patenting
dissemination decision itself. We discuss two types of societal influences: institu-
tionalized long-lasting influences and short-term random effects.
Long-lasting social change affects academic research in a slow and gradual way.
Social values, especially social values toward scientific research, can have long-last-
ing effects on scientific research decisions. For example, in the 1940s, Bush’s report
on science policy elevated basic research to a high priority in the scientific com-
munity. As a result, university researchers still tended to orient themselves more
toward basic research.
Societal forces also influence research through strong random shocks that ripple
through a society. Crisis and social disaster can influence the research agenda
directly. After “9-11,” security issues became a focus of both research ideas and
research funding. More scholars entered or shifted to this area of research, partially
as a result of increased funding allocations.
Because it provides the major portion of funds for university research, the federal
government is a major conduit for of both long-lasting and random influences in
academic research (National Science Foundation, 2002a). Social norms are
enhanced and transferred to universities through federal research funding and law.
The federal government supported growing basic research activities in universities
after World War II (Smith, 1990; Shapley & Roy, 1985). Similarly, during the 1980s,
concerns about international economic competitiveness resulted in several policies
aimed to boost applied research and technology transfer.
University Institutions Influences on Individual Decisions
Universities modify instructional factors (that is, academic focus, research foci,
institution setting, and so on) in order to maintain and increase their research
capacities (Wodarski, 1990). During this process, they generate institutional influ-
ence on research decisions.
Historically, many public universities in the United States were founded to sup-
port agriculture technology diffusion through education. Thus, teaching was the
initial academic focus of these universities. In the late 19th century, the first aca-
demic revolution made research an additional function of teaching universities
(Storr, 1952; Jencks & Reisman, 1968). More recently, a second academic revolution
has slowly transformed universities since the early and middle 20th century, toward
a tri-partite mission of teaching, research, and economic development (Etzkowitz,
2002). Each of these institutional transformations has directly affected individual
research decision making at all stages of the process.
Structural institutional changes also influence research decisions. A case study of
Columbia University shows dramatic increases in the number of patents assigned
to the university after the development of a technology licensing office in respond-
ing to the Bayh-Dole Act (Mowery et al., 2001). Research-facilitating policies (that
is, time release policy, funding support policy) provide institutional encouragement
for researchers to shift their research areas and methods of diffusing results. Struc-
tural changes in universities can also influence research direction. For example,
establishing new departments and institutions after World War II influenced the
research foci of universities (Shapley & Roy, 1985).
Societal Influence on University Institutional Factors
In addition to direct links between either society and the researcher or the univer-
sity and the researcher, societal influence also affects university institutions. Mow-
ery and Ziedonis (1999) identify a series of social and institutional forces causing
The Influence of Institutional Forces on University Patenting / 587
an increase in patenting and licensing at U.S. research universities after 1980. For
example, in order to make university research more relevant, the federal govern-
ment passed the Bayh-Dole Act of 1980, empowering universities to patent and
license research results funded by the federal government. The number of univer-
sity-sponsored research offices developed after passage of the Act is evidence of uni-
versity institutional change (Mowery et al., 2001). Similarly, the Supreme Court
decision Diamond v. Chakrabarty enabled the patenting and licensing of organisms
and molecules, thus creating a surge in biomedical research that resulted in
increased university patenting in this field.
Societal influence also manifests itself through the increased presence of research
centers at universities. Such centers typically receive money and personnel from a
combination of industry, federal government, and university sources. Centers have
multiple influences on university research. These centers are designed to fit the
practical needs of non-academic supporters, especially industry (Parker, 1997), and
thus place more emphasis on future/potential economic benefit derived from the
research. Also, researchers in these centers are sponsored by different resources,
thus expanding social relations and social networks (Rogers & Bozeman, 2001) to
work on specific technical problems. University researchers who work within this
environment are naturally influenced by such organization priorities and pay more
attention to the applied side of their research. On the other hand, the existence of
centers changes the traditional structure of the university, making it more difficult
for administrators to manage research outcomes from those centers (Bozeman &
Boardman, 2003). Because many of the societal pressures that influenced the rise
of these centers also influenced Bayh-Dole and funding decisions, we cannot sepa-
rately identify the effects of research centers in this paper.
Combining the theories of individual research production and societal influences,
the model in Figure 2 suggests four ways that outside changes might influence dif-
fusion decisions. First, social institutional change affects the diffusion choice
directly. Second, social change affects diffusion through changing university norms.
Third, policy responses to social change may affect research funding reallocation
and then influence the diffusion decision in turn. Fourth, university institutional
change influences the diffusion decision.
This framework allows us to consider general social forces as well as policy and
institutional decision process. In our empirical work, we develop proxy variables
for various short-run and long-run societal forces that influence the research
process. Because these social forces not only influence the patenting decision
directly, but also indirectly influence the patenting decision via research funding,
we model both university patents and applied research funding as endogenous vari-
ables. Thus, our work allows us to examine both how research funding and research
diffusion choices have been affected by policy.
While our theoretical framework analyzes an individual scientist’s decision
process, the unit of analysis of this research is aggregated research outcomes of a
university, rather than research outcomes of an individual researcher. The aggre-
gate number of patents granted to a university is a result of these individual scien-
tist decisions, along with decisions made by teams of researchers in research labo-
ratories or centers. Conceptually, the university and societal pressures affecting
such teams should be similar to those affecting individuals.
Three direct influences affect the ultimate decision to patent research results—
financial support of the research, university institutional characteristics, and socie-
588 / The Influence of Institutional Forces on University Patenting
tal forces. Since funding sources are influenced by many societal and university
institutional factors that affect university patenting at the same time, we model fed-
eral applied funding and industrial funding endogenously and propose the follow-
ing system of equations:
(1) Patents = f(Funding, University Institution, Social Forces)
(2) Federal Applied Funding = g(Social Forces, Economic Factors)
(3) Industrial Funding = h(Social Forces, Industry Income, University Institution)
In equation (1), funding controls for the effect of different funding sources on the
decision to patent university research. We look at three separate sources of funding:
federal basic research funding, federal applied research funding, and industrial
funding of university research. Equation (1) also controls for both institutional and
social forces that influence patenting decisions. Below, we discuss the role of
research funding on patenting. We discuss social and university forces in the fol-
lowing section.
Federal basic research funding, by definition, comes from the federal government
and supports basic research activities in universities. Since little basic research
leads directly to commercial applications, patents are an unlikely diffusion outcome
of basic research. However, basic research also serves as the building blocks for
future applied research. The more basic research contributes to the public knowl-
edge pool, the more productive applied research should be. Of course, to make a
basic research result commercially viable requires further applied research and
development, so that the effect of basic research is delayed. Therefore, we use one-
year lag data in our empirical model.
Unlike basic research funding, federal applied research to universities is intended
to produce research that meets a certain technological need. Commercialization is
often, although as noted earlier, not always, a goal of such research. When com-
mercialization is a goal, university research is typically just the first step in a longer
process of technology transfer. Because levels of federal applied research funding
are influenced by many of the same social factors that influence university patent-
ing decisions, it is endogenously determined in the model.
Equation (2) identifies long-run and short-run social forces as well as economic con-
ditions as the main determinants of federal applied research funding. In this research
context, long-run social forces track the overall influence changes in society’s prefer-
ences have on research, such as the focus on basic research after the Bush report.
Short-term social events also trigger ripples in policies effecting funding allocation pri-
orities. Examples include increases in energy R&D in response to higher energy prices
in the 1970s. Finally, the economic cycle is another important aspect that influences
research funding. When the economy weakens, government policy tends to encourage
technology development in hopes of spurring economic development.
Similarly, industry’s support for university research is also influenced by social
and economic forces, as presented in equation (3). For example, increases in indus-
try income, which determines the basic pool of resources available for industry to
contribute to R&D activities, should spur additional industrial research support.
Moreover, the industrial sector reacts to government policy, such as the amount of
federal applied funding. In order to capture the most benefit from R&D investment,
industry may attempt to take advantage of the free rider benefits of federally funded
applied research at universities, or from early involvement with new technologies
by following the research foci the federal government set for the nation. Therefore,
we expect general social forces, university level policies, and federal policies to
determine industrial applied funding along with aggregate business activity.
The Influence of Institutional Forces on University Patenting / 589
To test how university patenting behavior is influenced by changing institutional
factors inside and outside universities, we use university patent application and
research funding data from 1970 to 1997. To implement the model described in
equations (1)–(3), we also develop empirical representations of the various social
and university forces that impact patenting and funding decisions. Below, we pro-
vide detailed descriptions of the variables used to estimate the model. Table 1 pres-
ents descriptive statistics.
University Patents
“Patent Application” represents successful patent applications of those patents
granted in the U.S. and assigned to universities. As is standard in the literature on
patenting, we sort the data by the year of application. Sorting by year of applica-
tion, rather than year of grant, removes variations in the length of time it takes for
individual patents to move through the examination process. Raw data are calcu-
lated from the National Bureau of Economic Research (NBER) U.S. patent cita-
tions data file (Hall, Jaffe, & Trajtenberg, 2001),5which contains detailed informa-
tion on the almost 3 million U.S. patents granted between January 1963 and
December 1999.6During this time frame, only granted patents were published in
the U.S. Thus, our data include only successful patent applications. Furthermore,
since not all patent applications filed in recent years will have been granted by 1999,
Table 1. Summary of variables by type.
Variable N Mean SD Min. Max.
Dependent variable
Patent application 28 1003.607 823.257 208 3185
Funding variables
Federal basic research 28 6521.206 2022.312 4394.01 11603.78
Federal applied research 28 1898.983 621.993 962.8 2932.476
Industry research 28 835.464 530.648 225 2036
Short-term society variable
Bayh-Dole Act 28 0.571 0.504 0 1
Energy crisis 28 0.357 0.488 0 1
Human Genome Project 28 0.286 0.460 0 1
Long-term society variable
Hearing 28 2.679 2.597 0 10
Accumulated number of univ-
ersities successfully patented 28 270.607 118.840 99 468
GDP 28 5583.925 1350.682 3578 8856.5
Industry Income 28 2718.786 1534.545 697.5 6448.5
5 Source:
6 In addition to data taken from the NBER Web site, we also use additional data on the type of assignee
made available to the authors. This allows us to identify which patents have been assigned to universi-
ties. We thank Adam Jaffe for making these data available. As these data were only available through
1996, additional searches of university patents were performed by the authors.
590 / The Influence of Institutional Forces on University Patenting
we scale the data to adjust for patents still under examination.7Because most
patents take at least two years to go through the examination process, we only
include data through 1997 in our sample. Figure 1 displays these data. The figure
includes both the scaled data and the actual patent applications observed in the
Funding Characteristics
Funding variables include federal support for basic university research (“FedBa-
sic”), federal applied university research funding (“FedApplied”), and industrial
investment for university R&D (“Industry”). Data come from the NSF (2002a). All
funding data have been converted into 1996 U.S. dollars, and federal basic research
funding is lagged one-year, as discussed in the previous section. Figure 4 presents
trends in each of these three funding sources.
University Institutional Factors
We develop a proxy measure, “UnivNumber,” to capture the changing attitudes of
university researchers toward patenting as legitimate outcome for their research.
This variable reflects the growing number of universities that have successfully
Figure 4. University research funding by source.
7 We do this by first calculating the average grant lag for patents in our university patent data set. From
this, we estimate the percentage of pending patents for each year, and augment the data by this per-
centage. We calculate this percentage using only patents assigned to universities, as the grant lag for uni-
versity patents is typically longer than for other patents (Popp, Juhl, & Johnson, 2004).
The Influence of Institutional Forces on University Patenting / 591
applied for and received patents by year. We calculate it by accumulating the num-
ber of universities represented in the cumulative number of patents granted through
that year. As more universities develop positive attitudes toward patenting, we expect
stronger institutional forces within universities that will encourage researchers to
choose patents as one possible diffusion choice. The creation of technology transfer
offices is an example of such institutional forces. We calculate the accumulated
number of universities who had filed patent applications each year during the period
from 1970–1997 using data from the NBER patent citations data file. We include this
variable in both equations (1), which looks at university patenting, and equation (3),
which looks at industry funding of university research. We include university insti-
tutional factors in equation (3) to examine possible tensions between federal
research and industry-funded university research created by incentives for industry
to act as a free-rider. For example, we can test whether increased university patent-
ing allows industry to obtain licenses, rather than fund university R&D directly.
Long-Term Social Forces
To model the impact of long-term social forces, we examine how these forces affect
policymaking through increased political awareness of national technology devel-
opment. We use the number of Congressional hearings about university technology
transfer issues as a proxy for such awareness. This observation of political activity
assumes increased governmental activity comes in response to societal pressure for
long-term policy change. The variable “Hearing” is the number of Congressional
hearings related to university technology transfer issues in each year. As the early
stage of the formation of any policy, Congressional hearings about university tech-
nology transfer suggest increasing political concern about this issue. The raw data
are calculated from Lexis-Nexis Congressional Universe online database.8As hear-
ings capture general attitudes toward applied research, we use “Hearing” as an
explanatory variable in both the federal and industrial applied research equations
(equations [2] and [3]).
To look at specific policy decisions pertaining to university patenting, we create a
dummy variable to capture the effect of the Bayh-Dole Act. The variable “BayhD”
takes on the value one between 1982 and 1997 to represent the years in which the
Act was in effect. Moreover, since the intent of Bayh-Dole was to accelerate applied
work and technology transfer of federally funded research performed by universi-
ties, the impact of Bayh-Dole on patenting at universities depends on the level of
federal funding for applied research. Thus, we also interacted the Bayh-Dole
dummy with federal applied research funding. Both variables are included in the
patent application equation (equation [1]). We also include the Bayh-Dole dummy
in the federal applied research funding equation, as it is an indicator of the impor-
tance placed on such research funding.
Short-Term Social Forces
To study short-term social forces, we select two specific events from the time
period 1970–1997 known to have had an influence on science policy. They are the
8We searched for the key words “university basic research,” “university technology,” “university indus-
try technology,” “university patent,” “university innovation,” “academic basic research,” “federal R&D
contract,” and “university industry cooperation” appearing between 1970–1999 in the Lexis/Nexis CIS
592 / The Influence of Institutional Forces on University Patenting
energy crisis of the 1970s and 1980s and the start of the Human Genome Project
in 1991. The dummy variable “EnergyC,” takes on the value 1 between 1973 and
1985 to represent the years of highest energy prices, and a dummy variable
“HumanG” is 1 between 1990 and 1997 to represent the Human Genome Project.
In each case, exogenous short-run shocks led to increased demands for applied
research funding. As a result, both dummy variables are included in the federal
applied R&D equation.
The human genome project is of particular interest because research on the
human genome has resulted in many industry-university collaborations. As such,
we include both the human genome dummy variable and an interaction between
the dummy variable and federal applied research funding in the industry research
funding equation, to examine whether the relationship between federal and indus-
trial research sources changed after the human genome project began. Finally,
because biotechnology is an important source of university patenting, and the
human genome mapping project occurs after passage of Bayh-Dole, we include the
human genome dummy variable in the university patenting equation.
Economic Factors
We include two measures of income in our empirical work. Gross Domestic Prod-
uct, “GDP,” provides a measure of the national wealth that is the base for R&D fund-
ing. We expect a strong positive relationship between GDP and federal applied
research funding. GDP data is taken from the Bureau of Economic Analysis, trans-
ferred into 1996 U.S. dollars. Similarly, industry income, “IndIncome,” captures the
financial health of industry, which should affect the level of funding made available
from industries to universities. We use data on the Table of National Income With-
out Capital Consumption Adjustment by Industry Group from U.S. Department of
Commerce, Bureau of Economic Analysis. Private industry here is defined based on
the 1942 Standard Industrial Classification (SIC). The raw data are converted into
1996 chain dollars.
Because the two interaction variables in the model include both endogenous and
exogenous variables, we use generalized method of moments (GMM) regression to
estimate the system of equations (1)–(3). In addition to the explanatory variables
included in the equations, we also use a dummy variable for the political party of
the President as an additional instrumental variable. Table 2 shows the details of
the results for each equation.9
University Patenting. Of greatest interest is the effect of Bayh-Dole on university
patenting. Here, we find that Bayh-Dole does lead to a significant increase in uni-
versity patents. In particular, note that federal applied R&D had a negative
(although insignificant) effect on university patents before Bayh-Dole, but a signif-
icant positive effect after Bayh-Dole.10 After controlling for other exogenous forces,
it takes approximately $1 million of federal applied R&D to induce a new university
9 Since we are dealing with time series data, including one lagged variable, there is the potential for auto-
correlation to influence our results. However, further investigation reveals this is not a problem. Cor-
recting for autocorrelation yields no significant changes in the coefficients and does not significantly
improved standard errors.
10 Note that, although the Bayh-Dole dummy is negative, the joint effect of the dummy and the interacted
effect with R&D is positive for every level of applied R&D funding since passage of Bayh-Dole.
The Influence of Institutional Forces on University Patenting / 593
patent once Bayh-Dole is in effect. Given that, on average, about $1.5 million of
R&D is spent per patent in the U.S., the magnitude of this increase is important.
Moreover, while the Bayh-Dole Act does increase university patenting, it does not
lead to more federal applied research funding, as the Bayh-Dole coefficient is
insignificant in equation (2). This result supports other papers finding that Bayh-
Dole did not spur more research activity, but simply led to more patenting of
research results (see, for example, Henderson, Jaffe, & Trajtenberg, 1998). Indeed,
one motivation for the Bayh-Dole Act was to encourage technology transfer to
industry in order to increase commercialization of federally funded research. The
positive effect of federal applied research funding on patenting from equation (1)
combined with the insignificant effect of Bayh-Dole on federal applied funds in
equation (2) suggest an increase in the percentage of applied research transferred to
industry following Bayh-Dole, but not an increase in the level of applied research
activity. That is, Bayh-Dole has served to increase technology transfer, but has not
led to increases in federal applied research funding.
Turning to other influences on university patenting, we find that federal basic
research funding leads to an increase in university patenting, although the effect
is statistically insignificant. University culture is important. In equation (1), uni-
versity culture is represented by the accumulated number of universities that have
chosen to patent. We find this effect to be positive and significant. The variable
“UnivNumber” grows by an average of 7.7 percent per year in our sample period.
Table 2. GMM regression results.
Equation Variable Estimate Standard Error T-statistic Pr |t|
University patent Intercept –955.6380 1253.8000 –0.76 0.4548
FedBasic1 0.3610 0.2101 1.72 0.1012
FedApplied –0.8805 0.7925 –1.11 0.2797
Industry –3.2652*** 0.8659 –3.77 0.0012
BayhD –2780.4900* 1395.4000 –1.99 0.0601
FA_BD 1.8762** 0.8293 2.26 0.0350
UnivNumber 12.4298** 4.9961 2.49 0.0218
HumanG 180.6276 106.0000 1.70 0.1040
Federal applied
funding Intercept –485.5560** 194.2000 –2.50 0.0204
Hearing 10.1935** 3.8271 2.66 0.0142
GDP 0.4179*** 0.0430 9.72 .0001
EnergyC –37.7130 45.3568 –0.83 0.4146
HumanG 91.7099 105.0000 0.87 0.3919
BayhD –120.4390 75.2889 –1.60 0.1239
Industrial funding Intercept –271.4300** 97.6234 –2.78 0.0112
FedApplied 0.4162*** 0.1002 4.15 0.0005
Hearing –5.2918 4.7383 –1.12 0.2767
INDIncome 0.4609*** 0.1135 4.06 0.0006
UnivNumber –3.4221* 1.6566 –2.07 0.0514
FA_HG –0.8394** 0.3805 –2.21 0.0386
HumanG 2162.5240** 969.8000 2.23 0.0368
* Statistically significant at the 10% level. ** Statistically significant at the 5% level. *** Statistically
significant at the 1% level.
594 / The Influence of Institutional Forces on University Patenting
Increasing the average level of universities generating patents by 7.7 percent gen-
erates 259 new patents, or 20.8 new patents for each new university. As there are,
on average, 3.7 patents per university in our sample, this suggests that changes in
university culture work at least partially through external effects that increase
university patenting throughout all universities. Thus, creating a culture in which
university patenting is widely accepted may be as important as simplifying the
process by which such patents are approved. In contrast, short-run social forces
are less important, as the coefficient on the Human Genome dummy variable is
Unlike federal research funding, we find a strong negative relationship between
industrial research funding and university patents. Increasing industry R&D fund-
ing for universities by $1 million generates 3.27 fewer patents. This unexpected
result deserves more discussion. In our theory, we assume that industrial funding
encourages universities to perform more applied research and in turn to generate
more commercial results (that is, patents). However, industry’s cooperation with
universities is a complicated process, making the relationship between industry
funding and university-issued patents complex. Although industry funding may
induce more applied university research, it is not necessarily the case that such
research will end with a university patent application. This is not inconsistent with
the goals of Bayh-Dole, as a university-held patent may not be necessary to ensure
technology transfer when industry is the source of funds for applied university
research. To consider this further, we need to compare a university’s patenting moti-
vation to industry’s motivation as well as to separate the applied research process
from the patent production process.11
Although we have shown that universities have an increased interest in patenting,
industry still has a stronger motivation to own patents. There are several reasons why
industry, rather than universities, will be more likely to own the patents resulting
from collaborative work.12 First, companies use patents to protect their exclusive
commercial gains, as well as to reap the benefit from selling licenses. However, since
universities do not produce commercial products, universities benefit from patents
by licensing them to the private sector. If industrial funding is already in place, the
need for a university to hold a patent to seek out licensing partners is reduced. Not
only does this increase the incentives for industry to hold the patents, but some fac-
ulty members may simply choose to abandon the right for patenting and “sell” the
result to industry to save time and energy. Second, when agreements between uni-
versities and industry are being negotiated, industry may exert greater influence
because of its position as the research sponsor. Third, intuitional difficulties some-
times prevent universities from patenting industry-sponsored research. Some univer-
sities are not able or willing to afford the patent application fee and the labor costs
needed for the application process. Universities need efficient patent management
offices to deal with complex applications and technical license management tasks.
Federal Applied Research Funding. As noted earlier, our results show that while pas-
sage of the Bayh-Dole did increase university patenting, it did not lead to more
applied research funding from the federal government. Instead, our results suggest
11 We thank Professor Shiu-Kai Chin of the New York State Center for Advanced Technology in Computer
Applications and Software Engineering (CASE) and Gina Lee-Glauser, Director of the Office of Spon-
sored Programs, both at Syracuse University, for helpful discussions on this topic.
12 It may also be the case that both industry and the university apply for a patent together as co-assignees.
However, the company’s name usually shows as the first assignee name. In the NBER patent citations
database, the assignee name is the name of the first assignee in patent file. Therefore, it is possible that
we underestimate the actual number of universities involved in co-applications.
The Influence of Institutional Forces on University Patenting / 595
that both the national political and economic climate play an important role in
determining the level of federal applied research funding. Not surprisingly, federal
applied R&D spending increases when the economy is stronger. We represent the
political climate by the number of hearings on technology policy. As expected, the
sign of this variable is positive and significant. Federal applied R&D funding
increases by 27.3 million when the number of Congressional hearings on technology
transfer doubles in a given year, confirming the notion that federal applied research
funding is one of the most useful implementation tools in science and technology
policy. Finally, note that social events like the energy crisis do not influence the level
of federal funding policy for university applied research directly. It may be that soci-
etal events change the distribution of funding, but not the level of funding available.
Such a result, while beyond the scope of this paper, would suggest that short-term
social forces can crowd out federal R&D dollars on other topics.
Industrial Funding. Turning to industry funding for university R&D, we find that
industry income, social forces, and federal applied R&D spending all have significant
effects. As expected, industry income is positively correlated with industry support
for university research. In general, federal applied R&D and university R&D serve as
complements, as the coefficient on federal applied R&D is significantly positive. One
dollar of federal applied R&D support leads to $0.42 of additional industry support.
However, as we discuss below, the interaction of federal applied R&D and the human
genome dummy variable suggests important limitations to this finding.
In the industry R&D regression, we use dummy variables for the human genome
project to represent social forces. The mapping of the human genome and result-
ing biomedical research is an interesting case, as it presents an excellent example
of the mixed basic/applied research discussed earlier. As the biotechnology field
has evolved, it has made more use of collaborations between industry and univer-
sity than most fields. The effects of the human genome project on industrial R&D
funding are mixed. The coefficient on the dummy variable is significantly positive,
suggesting that the level of industry research support is responsive to short term
needs. Industry funding for university research increases by over $2 billion per
year during the Human Genome project. However, the interaction between federal
applied R&D and the human genome is negative—after the human genome proj-
ect began, federal applied R&D goes from being a complement to a substitute for
industry R&D. After the human genome project begins, a dollar of federally funded
applied R&D replaces $0.42 of industry-funded university R&D. A look at Figure 4
suggests that this result comes not from federal R&D crowding out industrial
R&D, but rather industrial R&D making up for slower growth in federal applied
R&D spending. In addition, the finding that the level of industrial R&D support
responds positively to such short-term social forces, whereas total levels of federal
applied R&D support do not, raises an interesting possibility. As noted above, it
may be the case that federal applied R&D responds not through increased levels of
R&D, but by changing the distribution of projects funded by the government.
While examining such changes is not possible with the data presented here, if this
were the case, shifting federal R&D support toward short-term applied needs
would simply take scarce research dollars from other federal programs, as
increased federal applied support would simply replace industry support.13
13 Popp (2002) finds similar results in the energy sector. There, he finds that government energy R&D
spending has a greater impact on private energy patenting activity after 1981, when the emphasis on
Department of Energy R&D spending shifted from applied projects such as synfuels to more basic
research projects
596 / The Influence of Institutional Forces on University Patenting
Finally, note that our university culture variable has a negative impact on industrial
support for university R&D. The same percentage growth in university culture that
leads to 259 new patents also leads to $71 million fewer industrial R&D support. This
negative relationship may indicate that a tension exists between industry and the uni-
versities over the ownership of patents. The more a university insists on owning the
patent, the less likely industry will continue its support of a research project. Another
possible explanation is that more university patents allow industry to use licensing as
a substitute for direct support of university research. In equation (1), we find that uni-
versity culture has a positive effect on university patenting. This increased university
patenting activity increases the possibilities for firms to license the results of univer-
sity research, rather than perform the research themselves. Indeed, encouraging such
transfers was one of the goals of Congress when it passed the Bayh-Dole Act.
The past 20 years have seen a remarkable growth in university patenting. This growth
has occurred within a policy climate placing increased emphasis on the commercial
relevance of government research. In this paper, we explore the reasons for increased
university patenting. Our theoretical framework considers both an individual
researcher’s motivation and institutional forces from inside and outside academia,
such as public policy. This framework allows us to investigate how governmental pol-
icy exerts both direct and indirect effects on individual decisions by university
researchers to patent applied research results. Like other work, we consider the
importance of legislation, particularly passage of the Bayh-Dole Act, which eased the
process of patenting the results of federally funded research. However, we also look
at forces that influence research funding decisions, as we jointly estimate the effects
of patenting and applied research funding from both government and industry.
Our results suggest that the Bayh-Dole Act has been successful in encouraging
transfer of technology from academia to industry. We find the Act increases the like-
lihood that federally funded applied research will result in a university patent.
Moreover, the Bayh-Dole Act does not lead to increases in the level of federal
applied research support for universities. Thus, patenting is not simply a result of
more applied R&D, but a result of more applied research results being patented. In
fact, despite the well-documented increase in the level of industry-supported
research at universities during the same time period, industry R&D actually leads
to fewer, rather than more, university patents! At the same time, our results suggest
that Bayh-Dole works not only by making it easier for universities to obtain patents,
but also by creating a culture in which university patenting is acceptable.
Also interesting is the small effect of short-term social events such as the human
genome project or the energy crisis. Industrial R&D support responds positively to
such social forces. However, federal applied R&D support does not. Clearly, such
shocks lead to changes in the distribution of federal R&D. For example, increases
in energy R&D during the 1970s are well-documented. That overall research efforts
do not increase suggests a possible crowding out effect—these short-term shocks do
not induce new federally funded research spending, but rather change the distribu-
tion of spending toward areas affected by the shock. Confirmation of such crowd-
ing out is left to future research.
Furthermore, the relationship between federal applied R&D funding and industrial
R&D funding changes when these social forces occur. During the human genome
project, federal applied R&D and industrial R&D acted as substitutes. This raises an
interesting policy question. A common justification for government R&D funding is
The Influence of Institutional Forces on University Patenting / 597
that it supports projects that would be underfunded if left to market forces. However,
we find that private R&D does increase in response to the human genome project. If
federal applied R&D simply replaces R&D support that would otherwise be provided
by industry, such funding may be unnecessary. Alternatively, while industry may
replace federally funded applied research dollars, the projects targeted may change.
As such, relying on industry support for R&D would change the nature of R&D
toward projects with greater commercial value. Given the increased importance of
industry funding at universities, such alternative explanations deserve further study.
Finally, we note that this paper presents the first research stage of research on insti-
tutional factors. We leave for future research a more detailed exploration of industry-
university cooperation. Of particular interest is what benefits these partnerships pro-
vide. Do universities benefit? Does the increased technology transfer observed through
more university patenting result in greater economic welfare? Other concerns may
focus on additional institutional factors that may influence university research. This
study focuses on macro-level influences, such as the economic and political climate.
Other social forces, such as knowledge and information exchange (Schartinger et al.,
2002), may also be important. Studying changes in behavior at the level of individual
universities would help us better understand the role of such forces.
YIXIN DAI is Research Associate, Center for Technology and Information Policy, The
Maxwell School, Syracuse University, Syracuse, NY.
DAVID POPP is Assistant Professor, Department of Public Administration, The
Maxwell School, Syracuse University, Syracuse, NY and Faculty Research Fellow,
National Bureau of Economic Research, Cambridge, MA.
STUART BRETSCHNEIDER is Professor, Department of Public Administration, The
Maxwell School, Syracuse University, Syracuse, NY.
Association of University Technology Managers. (1998). AUTM licensing survey: Survey sum-
mary. Norwalk, CT: Author.
Audretsch, D., Link, A., & Scott, J. (2002). Public/private technology partnership: Evaluating
SBIR-supported research. Research Policy, 31, 145–158.
Bozeman, B. (2000). Technology transfer and public policy: A review of research and theory.
Research Policy, 29, 627–655.
Bozeman, B., & Boardman, C. (2003). Managing the new multipurpose, multidiscipline uni-
versity research centers: Institutional innovation in academic community. Report for IBM
Center for the Business of Government.
Bush, V. (1980). Science—the endless frontier: A report to the President on a program for
postwar scientific research. Washington, DC: National Science Foundation.
Cole, J., & Cole, S. (1973). Social stratification in science. Chicago, IL: University of Chicago
Etzkowitz, H. (2002). MIT and the rise of entrepreneurial science. London: Routledge.
Etzkowitz, H. (2003). Research groups as “quasi-firms”: The invention of the entrepreneur-
ial university. Research Policy, 32, 109–121.
Hall, B., Jaffe, A., & Trajtenberg, M. (2001). The NBER patent citations data file: Lessons,
insights and methodological tools. National Bureau of Economic Research working paper
598 / The Influence of Institutional Forces on University Patenting
Harris, K. (1984). Factors associated with external funding in higher education: A review of
research. Journal of the Society of Research Administrator, 15(3), 39–47.
Henderson, R., Jaffe, A., & Trajtenberg, M. (1998). Universities as a source of commercial
technology: A detailed analysis of university patenting, 1965–1988. Review of Economics
and Statistics, 80(1), 119–127.
Jaffe, A., & Trajtenberg, M. (1996). Flows of knowledge from universities and federal labora-
tories: modeling the flow of patent citations over time and across institutional and geo-
graphic boundaries. Proceedings of the National Academy of Sciences of the United States
of America, 93(23), 12671–12677.
Jencks, C., & Reisman, D. (1968). The academic revolution. New York: Doubleday.
Link, A. (1996). On the classification of industrial R&D. Research Policy, 25, 397–401.
Moore, K. (1996). Organizing integrity: American science and the creation of public interest
organizations, 1955–1975. American Journal of Sociology, 101(6), 1592–1597.
Mowrey, D., Nelson, R., Sampat, B., & Ziedonis, A. (2001). The growth of patenting and
licensing by U.S. universities: An assessment of the effects of the Bayh-Dole act of 1980.
Research Policy, 30, 99–119.
Mowery, D., & Ziedonis, A. (1999). The effects of the Bayh-Dole Act on U.S. university
research and technology transfer: Analyzing data from entrants and incumbents, Meeting
of the Science and Technology Group.
National Science Foundation (NSF). (1996). Science and Engineering Indicators 1996.
National Science Foundation (NSF). (2002a). National Patterns of R&D Resources. Volume
National Science Foundation (NSF). (2002b). Science and Engineering Indicators 2002.
Owen-Smith, J., & Powell, W. (2001). To patent or not, faculty decisions and institutional suc-
cess at technology transfer. The Journal of Technology Transfer, 26(1–2), 99–114.
Parker, L. (1997). The Engineering Research Centers (ERC) program: An assessment of ben-
efits and outcomes. National Science Foundation report. Retrieved January 5, 2005, from
Popp, D. (2002). Induced innovation and energy prices. American Economic Review, 92(1),
Popp, D., Juhl, T., & Johnson, D. (2004). Time in Purgatory: Determinants of the Grant Lag
for U.S. Patent Applications. Topics of Economic Analysis & Policy, 4, Article 29. Retrieved
January 5, 2005, from
Rogers, J., & Bozeman, B. (2001). Knowledge value alliances: An alternative to R&D project
evaluation. Science, Technology, and Human Values, 26, 23–55.
Salter, A., & Martin, B. (2001). The economic benefits of publicly funded basic research: A
critical review. Research Policy, 30, 509–532.
Schartinger, D., Rammer, C., Fischer, M., & Frohlich, J. (2002). Knowledge interactions
between universities and industry in Austria. Research Policy, 31, 303–328.
Shapley, D., & Roy, R. (1985). Lost at the frontier—U.S. science and technology policy adrift.
Philadelphia, PA: iSi Press.
Smith, B. (1990). American science policy since World War II. Washington DC: The Brook-
ings Institution,
Storr, R. (1952). The beginnings of graduate education in America. Chicago, IL: University of
Chicago Press.
Thursby, J., Jensen, R., & Thursby, M. (2001). Objectives, characteristics and outcomes of
university licensing: A survey of major U.S. universities. Journal of Technology Transfer,
26, 59–72.
Wodarski, J. (1990). The university research enterprise. Springfield, IL: Charles C. Thomas
Reproducedwithpermissionofthecopyrightowner.Furtherreproductionprohibited without permission.
... For example, the United States, which has the objective to contribute to innovation in the industrial sector and take advantage of the investigations of the universities, created the previously mentioned law Bayh-Dole, which helps universities to transfer technology to the industrial sector through licenses. This law had positive results (Dai, Popp & Bretschneider, 2005) and helped the university to patent its investigations and transfer the technology in an effective way (Scott, 2004)for example there was an increase of spin-off technology created in the university (Mowery, 2011). Although the aforementioned law brought many benefits alongside it, it is important to be careful in applying these politics due to the reason that the investigators can look to distinguish their investigations, but only to receive benefits without dealing with the fundamental objective of the university: the education (Scott, 2004). ...
... Besides this, the investigations which are transferred more often, relate to disciplines within engineering and science (Hughes & Kitson, 2012) and come from basic investigations or applied developments in the university. Dai, Popp and Bretschneider (2005) explain that the university develops basic and applied investigation and that the amount of the investigation is influenced according to the environment in which it is found (society). Nilsson, Rickne and Bengtsson (2010), support the study and present a classification of the motives for which an investigator would transfer their knowledge to the industry. ...
... In its way, the state is creating politics which facilitate the creation of innovation funds, to help with the entrepreneurships and the technological research. Such politics also prefer the investment by means of the industry (Dai et al., 2005;Scott, 2004) and the impulse of the infrastructure (Saetre et al., 2009) -such as specialised branchesand the connections of the external suppliers (Khakbaz, 2012). At the same time, this results in the improvement of the investigation of culture and entrepreneurship, due to the politics being closely tied with the culture (Botelho & Almeida, 2011). ...
Conference Paper
Full-text available
Since the Bayh-Dole Act was enacted by the United States in the 80's, several countries in Latin American have developed national policies on technology transfer for the university context. For instance, this is the case of Mexico with the latest science and technology law enacted in 2002 and Brazil with the innovation law enacted in 2004, among others. In these laws are stated that the universities must organize technology transfer offices to manage intellectual property and the corresponding relationship with the enterprises. More recently, in Peru was enacted the new university law in 2015. This law includes orientations for the universities about the internal organization of technology transfer and intellectual property. In this context, the purpose of this study is the literature review about the identification and characterization of the different elements of the technology transfer processes between the university and enterprise. Also, the study aims to learn how the government of different countries is helping to foster the relationship between the university and enterprise through the technology transfer offices. For this has conducted a systematic review of five databases (EBSCO, Emerald, ProQuest, IEEE and Science Direct) and a repository of conferences of the Latin American Association of Technology Management (ALTEC). Firstly, in this search it was selected the papers with the technology transfer words in the title. The period of the searching was from the year 1980 to 2015 and the result was a total of 320 scientific papers in journals and 30 conference papers. Then, it was made an initial filtering with the keywords political, negotiation, barrier, international, strategy and model. Also it was made a second filtering with the keywords University, industry and government. The analysis of the gathered information in this study shows the evolution along the time of the several elements related with the technology transfer process in the university context. Thus, it is stated that there are a growing importance of the technology transfer offices and the university spin-offs. Finally, the literature review is important for outline a framework of policies for the context of Peruvian universities about the organization of the technology transfer.
... From this lens, and for the purposes of this study, industry and government partnerships and collaborations in the university tech transfer arena would be a part of a university tech transfer supply chain network. This is particularly true since industry and government sponsored research is driven by end-user's decision processes (Dai, Popp & Bretschneider, 2005). ...
Full-text available
There is a concentrated number of potential women entrepreneurs of diverse races among faculty in the United States' Historically Black Colleges and Universities (known as HBCUs and are called 'Black Colleges' herein). This study describes the potential for developing university technology transfer in these Black Colleges as a strategy for increasing diversity among women entrepreneurs in high growth, high tech fields using female academic entrepreneurs. Currently, Black Colleges lag behind their peer non-Black Colleges in. technology transfer because historically they have been under, served and were originally established largely as teaching and blue-collar trade schools. Although Black female STEM faculty comprised less than 2% of the US faculty, they are 22% at HBCUs (Mack, 2011). Using a novel theoretical framework, 24 Black Colleges with doctoral programs were compared to five (5) non-Black Colleges' technology transfer programs. The results of a correlation analysis support hypotheses regarding the relationships between tech transfer resource inputs and outputs.
... Therefore, in order to make the basic research results suitable for commercialization, further application research and development is essential. When commercialization is the final goal, scientific research in universities is usually only the first step in the long process of technology transfer [13]. That is, in the applied research stage, we conduct application research leading to secondary innovation to determine the possible use of the basic research results and to acquire new knowledge. ...
Full-text available
The innovation value chain is an effective tool for analysing innovation activities and reflects the process of value creation and increase in innovation activities. From the perspective of innovation value chains, we divided patent innovation activities into three stages: knowledge innovation stage, applied research stage and patent commercialization stage. The panel data from 64 universities directly managed by the Ministry of Education from 2009 to 2017 were used and several conclusions were drawn: 1) In the initial stage of knowledge innovation, the fundamental research fund plays a crucial promoting role, and knowledge innovation achievements are mainly published academic papers. 2) In the applied research stage, the knowledge innovation in the early stage and the fund investment in R&D activities have a significant positive effect on the patent output of universities, but the personnel investment has a negative effect. 3) In the final stage of patent commercialization, preliminary research results have a positive impact on patent commercialization, whose marginal effect depends on the industry-university-research relationship, external competition and reputation of the university. The evidence showed that there is a feedback channel between university patent commercialization and knowledge innovation, and new knowledge generated by the interaction with the outside world in the process of patent commercialization was transmitted to the subject of knowledge innovation through this channel, forming a virtuous dynamic cycle. By analysing the driving factors of the value chain of patent innovation in colleges and universities, we provided empirical evidence for the operation mechanism and policy formulation of college patents in China.
... The Bayh-Dole Act helps the university in the process of transferring technology to the industry through intellectual property. The Bayh-Dole law obtained positive results (Dai, Popp & Bretschneider, 2005) and helped the university patent its research and transfer Proceedings of the International Conference on Industrial Engineering and Operations Management Bogota, Colombia, October 25-26, 2017 © IEOM Society International technology effectively (Scott 2004). A clear example was the spin-off increase with technology base created in the university (Mowery, 2011). ...
Conference Paper
Full-text available
Over the years, one of the most important problems in engineering university education is the lack of student attention. The great majority of this problem is that the student feels that the knowledge transmitted in class has nothing to do with the real world. Besides that, the industry needs creative and innovative solutions and universities are best suited to meet the challenge through attractive and relevant curricula integrating innovation topics. For this reason, there is a phenomenon has emerged, known as collaboration between the university and industry (CUI), which becomes very beneficial for both agents. This article aims to show a case study of a collaboration project between small and medium-sized enterprises (SME's) and the entrepreneurship of a team of teachers and students. The project, called MicroErp, is being developed through end-of-course projects of the computing faculty and consists of the creation of a computer tool for medium and small business resource management.
... In the USA, the regulatory framework for patenting in universities is considered to be the Bayh-Dole Act of 1980 (Henderson et al., 1998;OECD, 2000). The number of patents granted to universities in the USA has increased significantly after the Bayh-Dole Act (Dai et al., 2005). ...
One of the roles to be played by the university, besides the production of knowledge, is its contribution to technological innovation, which in some cases is manifested by the filing of patents. This characteristic becomes more important in countries that are in the development stage, as is the case of Brazil, one of the members of the BRICS. In this context, the following questions arise: what does the patent filing map in Brazil with the participation of Brazilian universities look like? And what are the factors that stimulate or inhibit the production of patents in this country? By identifying the main inhibiting and stimulating factors for the patenting in Brazilian universities from the perspective of innovation agencies, this unprecedented study may support policies to promote technological innovation and, in particular, to increase patent filing with the participation of universities. Suggested Citation: Fernanda De Carvalho Pereira & Helder Gomes Costa & Valdecy Pereira, 2019. "Stimulating and inhibiting factors of patent filing with Brazilian universities," International Journal of Entrepreneurship and Innovation Management, Inderscience Enterprises Ltd, vol. 23(3), pages 261-280. DOI: 10.1504/IJEIM.2019.099844.
Full-text available
Purpose - Due to the frenetic and dynamic working conditions ascribed to Architecture, Engineering and Construction (AEC) project organizations, enormous research has addressed the poor mental health propensity of project management practitioners (PMPs). However, research has not considered the distant factors related to organizational design causing poor mental health. Therefore, this study addresses the problem by integrating institutional theory, agency theory and resource-based theory (RBT) to explore the relationship between organizational design elements: project governance, knowledge management, integrated project delivery, project management skills and mental health management indicators. Examples of mental health management indicators include social relationships, work-life balance and project leadership. Design/methodology/approach- Purposive sampling method was adopted to collect survey data from 90 PMPs in 60 AEC firms in Australia. Structural equation modelling (SEM) was utilized to test the relationship between the variables. Findings - The research found that project governance, knowledge management and integrated project delivery are positively correlated to mental health management indicators. However, the research finding suggests that project management skills have a negative impact on mental health management indicators. Originality/value- The findings offer guidelines to AEC firms on achieving positive mental health management outcomes through concentration on project governance, knowledge management and integrated project delivery. It further calls for a reconsideration of existing project management skills causing poor mental health management outcomes.
The article overviews the existing models of technology transfer, including those within foreign universities, and highlights the most relevant ones that can be used by Russian universities in the post-COVID-19 conditions. The study should allow the university-based transfer centers to choose the model which is mostly suitable for their situation, and to include elements that will help them to maximize the efficiency of their activities. The existing centers will be able to make changes in their activity in order to update and/or to transform it in accordance with the changed conditions. For the management personnel of the university, the article also provides practical recommendations on managing technology transfer centers. The authors reveal the key functioning elements of various technology transfer models, which can be used by management personnel to design technology transfer centers based on Russian universities. The possible result of the stakeholders’ getting to know this study might be their creating and implementing regulations to govern the technology transfer centers’ activities; forming a personnel reserve; advanced existing personnel training and multi-competence teams’ creating; forming a flexible budgetary policy, as well as a policy of values, for the technology transfer center to function within.
The Bayh–Dole Act was enacted in the United States in 1980 to promote economic development and growth at regional and national levels. A key engine is research generated within universities. This article addresses the question of how universities can serve as engines of development. Drawing on Cooter and Shaeffer’s work on law and development, specifically what they call the double trust problem, this article shows how the Bayh–Dole Act was justified as resolving the double trust problem arising from lack of property rights in university research. This article presents the argument that this goal of the Bayh–Dole Act ignores how universities solve another dimension of the double trust problem, namely the generation of human capital. The author examines the theoretical justifications for the Bayh–Dole Act and universities and the empirical policy literature assessing university patenting and commercialization in the United States, South Africa, and India.
In this mixed-methods study, we examined student and administrator perceptions concerning student-created intellectual property (IP) ownership in the higher education setting. Survey results showed students had low levels of IP-related knowledge and worried that the university might claim ownership of their ideas or class projects. Students also reported inadequate communication from the university concerning IP policy rules and technology transfer processes. Interviews with campus IP administrators about student survey results led us to assert that communication inadequacies are perpetuating misunderstanding between students and university IP ownership intentions. Recommendations include implementation of several proactive communication activities and an IP training program for faculty.
Full-text available
The question of what the relevant entities or units of analysis for studying the dynamics of R&D are is central not only for adequate characterizations of the system of scientific and technological knowledge production but also for determining the correct focus for evaluation of R&D activities. Typically, R&D performance evaluations have focused not only on the wrong thing but have looked in the wrong place. Most evaluations have been project or program based. Often this focus is misleading. This article presents a “knowledge value” framework as an alternative focus for understanding and evaluating scientific and technical work. This framework consists of two core concepts: the Knowledge Value Collective (KVC) and the Knowledge Value Alliance (KVA). On the basis of the analysis of twenty-eight case studies of research activities, the authors present a typology of KVAs and conclude that they are a better object of evaluation than discipline-based projects.
Just after the close of World War II, America's political and scientific leaders reached an informal consensus on how science could best serve the nation and how government might best support science. The consensus lasted a generation before it broke under the pressures created by the Vietnam War. Since then the nation has struggled to reestablish shared beliefs about the means and goals of science policy. In American Science Policy Since World War II, author Bruce L. R. Smith makes sense of the break between science and government and identifies the patterns on postwar science affairs. He explains that what might otherwise seem to be a miscellaneous set of separate episodes actually constituted a continuing debate of national importance that was closely linked to broad political and economic trends. Smith's precise and unique analysis gives both the scholar and historian a better understanding of where we are and how we got there while casting a modest light on future policy directions.
This article is concerned with military security for war of science against disease. The research in medicine and biology should be continued.