Objectives, agreements and matching in science–industry collaborations: reassembling the pieces of the puzzle Nicolas Carayol (2003) OM 석사과정 이혜성.

Presentation on theme: "Objectives, agreements and matching in science–industry collaborations: reassembling the pieces of the puzzle Nicolas Carayol (2003) OM 석사과정 이혜성."— Presentation transcript:

1 Objectives, agreements and matching in science–industry collaborations: reassembling the pieces of the puzzle Nicolas Carayol (2003) OM 석사과정 이혜성

2 1 Introduction 과학 분야와 산업 분야의 연구개발에는 어떠한 공통점과 차이점을 생각해볼 수 있을까?

3 Science-industry collaborations' impact on economic performances
1 Introduction Gap: Science–industry collaborations have been the subject of a considerable attention. However, the existing studies are still restricted to a partial view of the phenomenon. Carayol (2003) studies for science-industry collaborations, relative to their characteristics, their aims and the collaboration settings. Science-industry collaborations' impact on economic performances Authors 1. most scholars highlighted their social benefits Brooks, 1994 1-1) increasing the performances of the collaborating firms, the productivity of their R&D processes David et al., 1992; Zucker and Darby, 2000 1-2) increasing absorptive capacity Cohen and Levinthal, 1989 2) looking at the outcomes arising on the other side of the collaborations, by improving the economic relevance of scientific knowledge production Gibbons et al., 1994; Gibbons, 1997 2. risks and costs expressed 2-1) in terms of inopportune modifications of public research agendas toward more applied research Cohen and Randazzese, 1996; David, 2000) 2-2) on knowledge disclosure restrictions they may generate Cohen et al., 1994; Blumenthal et al., 1997

4 1 Introduction What science–industry collaborations really are?
First, knowing the determinants of each objective: can we find characteristics of partners that are correlated to the stressing of specific aims? Secondly, which academics’ aim “fits better” with which firms’ aim and vice versa? Finally, what are the different collaboration agreements which are used to simultaneously serve these objectives? Hypothesis: “assortative” matching process - An “assortative” matching process is said to be arising when best ranked agents (on a given criterion) preferentially match together (for our purpose collaborate).

5 1 Introduction Data: data on the collaborations and on the partners.
- Original data: within the SESI-TSER project network3 in five countries (Austria, France, Germany, UK, US). Interview firms: information and telecommunication (IT) and pharmaceuticals and health related biotechnologies (Ph/Bio) sectors and some of their academic partners.

6 2 What do we know about science–industry collaborations?
2.1. The various forms and intensity of interactions 2.2. Contractual agreements 2.3. Academic partners’ objectives for collaborating 2.4. The “unintended consequences” 2.5. Surveying the firms side

7 2 What do we know about science–industry collaborations?
2.1. The various forms and intensity of interactions Meyer-Krahmer and Schmoch (1998) a survey of more than 400 scientists “representing” German research centres belonging to five fields The more applied fields are the ones that receive the more important share of their funding from industry. Schartinger et al., 2002 to measure the sectoral pattern for different types of knowledge interactions and explored the determinants of interactions at the level of scientific fields and sectors

8 2 What do we know about science–industry collaborations?
2.2. Contractual agreements Cassier (1997) 150 contracts of laboratories of the French CNRS the institutional settings partners agreed on Joly and Mangematin (1996) a large French research institution specialised in agricultural research (INRA) & their collaborations with firms three “logics” for collaborating: a “geographical proximity logic”, a “market oriented logic” and a “symbiotic”

9 2 What do we know about science–industry collaborations?
2.3. Academic partners’ objectives for collaborating Thursby et al. (2001) data from the Association for University Technology Managers (AUTM) survey and built their own survey of 62 TTO managers royalties generated are lower and sponsored research is more likely when the new technology is licensed at an early stage of development Meyer-Krahmer and Schmoch (1998) advantages and disadvantages of laboratories for interacting with industry Lee (2000) Surveying more than 100 academic scientists in the US reasons for collaborating & the perceptions academics have of their gains from collaborating with industry

10 2 What do we know about science–industry collaborations?
2.4. The “unintended consequences” Berens and Gray (2001) “unintended consequences” Dasgupta and David, 1994; Stephan, 1996 the efficiency properties of collective knowledge production relying on the norms and rules, tend to be attentive to the various risks science–industry collaborations Cohen et al. (1994) data collected surveying directors of US universities University Industry Collaborative Research Centres (UICRC). They first evidenced that collaborating with industry implied restrictions to publication, and was correlated to performing applied research. Cohen et al. (1998) UICRC aim to “improve industry products and processes” Thursby and Thursby (2000) patenting and licensing activities in the US universities the issue whether the dramatic increase in the patenting and licensing activities in the US universities was due to an increase in the willingness to patent discoveries or if it was the consequence of a shift in the nature of their research contents. Gluck et al. (1987) surveying more than 700 Ph.D. candidates at life science departments of major US universities. if the Ph.D. or their supervisors were involved in science–industry collaborations, then their publishing and disclosure behaviours were altered. Blumenthal et al. (1996) surveying more than 2000 public researchers in the life sciences. Scientists' productivity, when they received industrial funds. Blumenthal et al. (1997). science–industry collaborations associated with data withholding.

11 2 What do we know about science–industry collaborations?
2.5. Surveying the firms side Adams et al. (2001) the firms involved in collaborations with the UICRCs. collaborating with academics is more a complement to their own research than a substitute, by evidencing that interacting with a UICRC increases R&D spending and the number of patents awarded. Hall et al. (2001) a limited set of contractual data from the US government ATP research program the probability that an IPR-related “insurmountable” barrier arose in the relationship was increased when the intellectual property characteristics of the research were certain and when the research was expected to lead to less appropriate results. Hall et al. (2000) ATP projects data, including now 47 cases, among which 30 involved universities The involvement of university researchers creates “research awareness” in the aims of the projects. Caloghirou et al. (2001) a survey of 285 cases of collaborative projects funded through EU framework programs the intensity of the collaboration with universities was positively correlated with the declared importance of aiming Lee (2000) survey of academics & 140 firms collaborating with universities by partnering with faculty members, firms were first “gaining increased access to new research and discoveries”, second making “significant progress toward the development of new products and processes”, and third helping them significantly toward a closer relationship with the university”. Zucker and Darby (1996) followed by Zucker and Darby (2000) biotech firms the consequences of either hiring or co-publishing with “star scientists” on the firms research productivity

12 3 The data

13 4 The diversity of science–industry collaborations: a typology
4.1. The methodology The formal data analysis aims to construct a typology of the relations between academics and firms. a multi-correspondence analysis (MCA) - 변수 식별과 dataset의 dimensionality 감소를 위해 (ii) an ascendant hierarchical classification - 동질적인 그룹으로 모수를 분리하기 위해 사용 & four axes에서 개인들(관계들)을 조정 (iii) a correlation analysis - 타입들과 진행되는 분석에 포함되지 않은 더 많은 구조적 변수들 간의 links를 구체화하기 위해 활용

14 4 The diversity of science–industry collaborations: a typology
4.2. The typology organisational solutions, consortium, volume, duration, novelty, nature and risk of the research Axe 1 the intensity of the basicity, the risk and the novelty of the collaborative research Axe 2 opposes the networked research to the bilateral one Axe 3 comes from the very originality of start-up creation compared to standard science–industry collaboration Axe 4 the distinction between collaborations that are very organisationally structured (mainly belonging to the second type) to the others.

15 4 The diversity of science–industry collaborations: a typology

16 4 The diversity of science–industry collaborations: a typology

17 5 The matching of academics and firms 5.1. Research agendas
science–industry collaborations 유형은 협력 연구의 본질이 파트너의 특성과 전략에 따라 다양함을 보여줌. 저자는 다양성을 설정하는 하나의 측면으로 agents들을 통합하는 matching process 제시 matching process: 상호 작용 시 agents들의 다양한 전략적 고려사항. Research agenda 의 개념이 아카데미와 기업 agents들의 전략에 중요하며 matching을 설명하는 중요한 요소. 어떻게 각 side의 이질적인 actor들이 연구 agendas의 전략적 정의로부터 협력을 고려하는지 시스템적 탐색. Research agenda: 연구 목적, 집합적 지식 생산이 공개된 행동을 이끌어 연구자들 간의 경쟁이 적절한 연구 목적을 선택하는 것. 학계 연구자: 연구에서의 Research agenda의 특성으로 기회비용과 이익을 고려하여 결정 산업 파트너: risk borne에 의존.

18 5 The matching of academics and firms
5.2. A simple rationale for the matching The academics 과학-산업 간 협력은 그들 자신의 연구 목적을 추구하는 것을 연기하는 비용을 반영한 기회 비용을 반영한다. 기회 비용은 자신의 연구 목적을 추구하는 시간과 노력을 연기하는 비용을 나타내며, 그대신 그 비용에 대응하는 협력연구에서 연구자의 의도에 맞는 연구 주제를 수립하도록 하는 기능을 한다. 아카데미는 기업과의 협력 기준으로 협력 연구와 그들 자신 연구 agenda 간의 fit을 고려하기 때문에, 아카데미와 기업 간 agenda의 distance가 클수록 시너지가 작을 것이라고 저자 주장. The firms 기업이 아카데미 파트너와의 협력과 research agenda를 선택하는 방식은 두 가지이다. (1) 위험을 최소화, (2) 혁신을 지속하기 위해 더 위험한 협력 연구 프로젝트를 개발한다. 저자는 아카데미의 제휴 기관의 명성과 그들의 연구의 basicity의 정도가 지속가능한 연구 능력의 신호로서 사용될 수 있다고 주장하였다.

19 5 The matching of academics and firms
An “assortative” matching: cash flows와 IPRs로 표현 아카데미 파트너의 협력 의도가 감소할수록 아카데미 파트너에 의해 요구되는 가격은 높아지고, 산업 파트너의 협력 의도가 높을수록 산업 파트너의 유보 가격(reservation price)는 높아진다. HH와 LL은 매우 전형적인 “assortative” matching configurations으로서 가장 높은 랭크 agents가 가장 높은 랭크의 것과 우선적으로 맞춘다는 가정

20 5 The matching of academics and firms
5.3. Testing the “assortative” match 아카데미 연구 agenda와 그들의 "주요 협력 목적"의 특성이 correlated되는지 체크 “assortative” matching hypothesis를 직접적으로 테스트 연구 결과, 아카데미 전략은 연구 agendas 특성 변수와 연관되어 있다. 협력 변수를 통한 아카데미 전략의 "증가된 과학적 우수성" 양상은 "기초 연구의 수행""높은 수준의 우수성"과 긍정적으로 연관. 이는 아카데미 연구소 연구 agenda의 본질이 협력 전략의 정의에 중요성이 있다는 견해를 지지한다. 매칭 프로세스는 아카데미 연구 우수성이 높게 신호되고, 연구가 기초적인 경향이 있을 때, 높은 수준의 위험을 지지하는 기업과 연관되는 것을 선호한다는 것을 보여준다.

21 6 Conclusion Research outcomes
(1) The first outcome of this study is a five set typology of collaborations obtained through a multi-correspondence analysis followed by an ascendant hierarchical classification. (2) The second outcome of the study is a rationale for understanding the matching of potential academics and firms based on the nature of the research the former perform and the nature of the research realised in common.

22 7 Discussion 본 논문은 산학 협력의 context에서 논의하고 있는데, 산학 협력이 대학 간이나 기업 간, 또는 연구소 등의 다른 주체를 포함한 것과 다른 점은 무엇인가? 본 연구의 분류를 적합하다고 생각하는가? 만일 그렇지 않다면 어떠한 기준을 제시할 수 있겠는가?

23 7 Discussion

24 감사합니다.