Parallel Session Details

Track A: Innovation Policy

A1: A Decade of Research on the Entrepreneurial University: A Bibliometric Review (Extended Abstract)

Martin Bliemel1, Angelo Cavallo2, Antonio Ghezzi2, Lorenzo Monicolini2

1University of Technology Sydney, Australia; 2Politecnico di Milan, Italy

The leading role of universities to be entrepreneurial is widely recognized by academics, businesses, government, and the third sector. Since Rothaermel, Agung and Jiang’s seminal review in 2007, the literature on the Entrepreneurial University experienced exponential growth, with over 1500 articles over 10 years. This paper aims at providing an objective, systematic, and comprehensive review of the literature based on detailed bibliometric analysis by building on the keywords used by Rothearmel et al (2007) and including new keywords derived from Wright, Siegel and Mustar (2017). Our review identifies seven key themes and provides a critical review of the underlying research streams, while noting their interconnectedness. This literature review lends evidence to the integration of entrepreneurship education into the literature on the entrepreneurial university. It also clearly identifies the articles and their authors that most epitomise each of the seven themes, as a starting point for scholars interested in each of the themes. We conclude the review with the limitations and future research opportunities regarding the Entrepreneurial University.

A2: The growth patterns of firms in the Korean Semiconductor Industry sector: Hidden Markov Model analysis (Extended Abstract)

Seokhyun Ryu, Sanghyun Park, Sungjoo Lee

Ajou University, Korea, Republic of (South Korea)

Semiconductor industry in South Korea has been steadily growing for the past 40 years, and is now the leading Korean industry that has achieved the highest level in the world. However, the semiconductor industry in South Korea has recently entered a period of great change due to the rapidly changing global industrial environment, such as the pursuit of China, the strengthening of Japan-Korea trade conflict and protectionism. In order to respond to the times of these major changes, the need to reilluminate the industrial development process from the perspective of the industry value chain was raised. Therefore, in this study, based on the Hidden markov model, we analyze the growth structure and growth patterns of the Korean semiconductor industry by utilizing the firm’s revenue data of materials ∙ parts ∙ equipment companies that belong to the area of the upstream. Our analysis results can provide insights into the Korean semiconductor industry in terms of the value chain and provide a basic framework for future semiconductor industry research.

A3: From Feedback to Feedforward: Reg-tech Application to Fraud Detection (Extended Abstract)

Joonmo Ahn, Changyong Lee, Seongwook Moon

Sogang University, Korea, Republic of (South Korea)

Recent years have witnessed increasing collision between technology and regulation, and this conflict has been extensively discussed by both scholars and policy makers. To date, ‘smart regulation’, such as innovation procurement, has attempted to reconcile conflict between innovation and regulation, but, thanks to recent new digital technologies, such as big data and artificial intelligent (A.I.), the opportunity of this co-evolution is increasing.  A good example of this intelligent evolution would be ‘Reg-tech’ which offers more advanced financial service while decreasing the costs for financial transaction monitoring using information and communication technology (ICT). This explorative study goes beyond fin-tech and attempts to apply machine learning technology for Workers’ compensation and employment (WCE) insurance fraud detection. The comparison of traditional expert based audit and four different manchine learning analysis results suggest that the proposed approach shows better performance. Our illustrative case suggests promising possibility of intelligent co-evolution between regulation and technological innovation. To date, regulation enforcement has been a feed-back process, in which a violation of regulation precedes and penalty for the violation follows. However, big data and accurate machine learning technology changes this traditional process from feed-back to feed-forward. The utilisation of machine learning technology can not only increase regulation enforcement productivity but also reduces compliance costs by signalling pre-preventive alarming to risky groups.

A4: Happily ever after? – The longitudinal impact of government R&D subsidy on innovative small firms (Extended Abstract)

Joonmo Ahn1, Tae Chung Kim2

1Sogang University, Korea, Republic of (South Korea); 2Sogang University, Korea, Republic of (South Korea)

The current paper casts doubt on (some of) the impacts of R&D subsidies policy on innovation. To date, many governments have just blindly tried to increase their innovation inputs without fully appreciating the consequences of these decisions. Recent studies have reported complicated influence of policy intervention (e.g., an inverted U-shape of policy impact) (Ahn et al., 2020), and this study adds additional evidence for this complex nature of innovation policy. Our empirical analysis results suggest that, (1) public subsidy may contribute to the enhancement of innovation input, output and behavioural changes in small innovative firms, but (2) this positive influence driven by R&D subsidy does not last long. Our results indicate that subsidy-based policy may not effective for the establishment of sustainable changes. This study implies that current subsidy-oriented policy must be renovated and carefully designed to consider longitudinal impact of policy intervention and sustainable changes for innovation.

A5: Open to fail: Collaboration and abandonment among innovating firms (Extended Abstract)

Mark Freel1, Dirk Czarnitzki2, Kristof Van Criekingen3

1Telfer School of Managment, Ottawa, Canada; 2KU Leuven, Belgium; 3Aarhus University, Denmark

Our analysis contributes to the understanding of the link between Open Innovation ("OI") and project abandonment. Scholarship has mostly advocated the benefits of open innovation, notwithstanding some notable exceptions. In this paper we find additional evidence for the darker side of open innovation as project abandonment and completion are both influenced by the  number of collaboration partner types. This evidence detracts somewhat from the mainly positive perspective that OI scholarship has taken. Our research is especially policy relevant given that it shows that the balance between OI  benefits and costs is not purely positive as it is often advocated to be. We also find that especially industry-university collaborations are prone to being exposed to potential OI drawbacks, meaning that especially in this type of setting some of the prevailing knowledge on OI is to be taken with adequate caution and indeed, could or should be re-evaluated.

A6: Godfathers and Dissemination: Knowledge Spillovers and Ecosystems in Artificial Intelligence (Extended Abstract)

Margaret Dalziel1, Xiao {Mimosa} Zhao2, Andrew Watkins3, Phil Shapira4

1University of Waterloo, Canada; 2University of Toronto, Canada; 3Manchester University, UK; 4Manchester University, UK, Georgia Institute of Technology

Some knowledge is easily disseminated, some with difficulty, and some resists transfer.  Despite the global knowledge sharing mandate of public science, there are locational effects to the exploitation of academic research.  This motivates governments to invest in institutes and programs designed to give new and established firms in the vicinity of leading scientists an advantage.  We consider the case of Geoffrey Hinton (University of Toronto), Yoshua Bengio (Université of Montréal) and Yann LeCun (New York University), who have been referred to as the godfathers of artificial intelligence.  We examine the factors that affect the ability of private technology developments to absorb and build upon their research.  Specifically we model the likelihood that an artificial intelligence patent cites the research of a godfather as a function of godfather-assignee relations, inventors that were former students or post-docs, co-inventors, or co-authors, and geographic proximity, and find significant positive effects for most social relations but a significant negative effect of geographic proximity.  While this paper is a work in progress, it may be that proximate firms lacked the capabilities required to capitalize on the locational opportunities.

A7: Innovation in the Cranberry Industry: A study of industry dynamics from the perspective of cluster theory (Extended Abstract)

Andy Hira

Simon Fraser University, Canada

Economic cluster theory tries to explain why competing firms co-locate. It has been applied to innumerable studies of concentrations of industrial corridors, from Silicon Valley to the financial districts of New York and London.  However, outside of some studies of the wine industry, a luxury product, it has not been extensively applied to agriculture. We tend to think of agricultural location decisions as based on climate, soil, and water. In this paper, we examine the seven clusters of cranberry growing in North America, and see that elements of industrial cluster theory apply. Proximity to R&D, tacit knowledge, shared personnel, and marketing all have an effect on decision-making beyond natural environmental factors. The research should open the way to re-examine agriculture in regard to the presence of important if underestimated innovation and coordination systems that help explain competitive advantage in regard to location.

A8: Knowledge reuse and capture from design patents: A case of a future mobility door (Extended Abstract)

Siyoung Yoon, Kisik Song, Sungjoo Lee

Ajou University, Korea, Republic of (South Korea)

Design-based strategies are becoming more dominant in corporate competition than technology-based strategies, and design patents are emerging as a mechanism for capturing technology opportunities. However, despite the potential of design patents as a source of technical information, most studies on capturing technology opportunities focus on utility patents. Despite some studies being conducted, these studies mainly use the image information of the design patent. Therefore, in this study, we propose a methodology to derive a design reference field using text information of a design patent and discover a design concept idea in the derived design reference field. In addition, we intend to verify the effectiveness of the methodology by applying the proposed methodology to a future mobility door. The methodology proposed in the study is expected to contribute to the development of a company's product design concept by discovering technology opportunities with reference to the design of other industries.

A9: Is the Chinese IP system fair? – An investigation of judicial local protection and judicial local support in IP litigations in Beijing and Shenzhen (Extended Abstract)

Ya-Feng Zhang1, Crupi Antonio2, Di Minin Alberto2, von Zedtwitz Max3, Hai-Bo Liu4

1School of Public Policy and Management, University of Chinese Academy of Sciences; 2Institute of Management - Scuola Superiore Sant'Anna; 3Department of International Economics, Government and Business, Copenhagen Business School; 4Institutes of Science and Development, Chinese Academy of Sciences

China is not only one of the world’s most productive intellectual property (IP) creators but also one of the most prolific IP litigators. In this context we apply an institution‐based view to investigate the fairness of IP courts in two Chinese cities, Beijing and Shenzhen. We introduce the concepts of judicial local protection (JLP) and judicial local support (JLS) to gauge fairness of IP system. The assumption is that local firms receive preferential outcomes from courts with whom they share the same ecosystem. However, our findings show that there is no preferential treatment, neither for JLP nor for JLS. There is even evidence that non-local litigators obtain more favorable results. Thus, we contribute to the discussion on the evolution of IP institutions in China in particular and developing economies in general. We also report on possible abuses of the IP system in China. Therefore, this paper also contributes to the literature on and practice of international business, local investment, and global collaborations.

A10: Principal Investigator (PI): 'Shock Absorber' for quadruple helix value co-creation (Extended Abstract)

Lawrence I. Dooley1, Conor O'Kane2, James Cunningham3

1UCC, Ireland; 2University of Otago, New Zealand; 3Northumberia University, UK

With the triple/quadruple helix model placing ever increasing pressure on universities to enhance its role in the National Innovation System (NIS). This has resulted in the scaling of fundamental research capability within university through Government backed Funding Agencies, to create a critical mass of potential innovation for exploitation.  This research explores the systemisation of traditional University research within large scale Research Centres and the role of Principal Investigator (PI) in aligning stakeholder objectives of the triple helix model for value congruence. Findings show that while increased funding and systemisation of the University Research base can be advantageous in building research capability within the National Innovation System (NIS), it also results in tensions that must be managed and places responsibility on research PI’s to behave as a shock-absorber to achieve the promised research synthesis and breakthroughs.