Pecs Session Week 2 — July 2009

Pecs Session / Week 2 - July 2009 STI policy rationales. Part I : paradigms Laurent Bach laurent. bach@unistra. fr BETA, university Strasbourg STI Policy rationales. Part I : Paradigms - L. BACH 1

Setting the scene… Public research infrastructure (universities, research centers, …) research activities incentives for researchers Technology procurement policy Purchase/pioneer use by public entities (administrations, organisms, public companies, Support to cooperation between firms and public research "valorization" - Tech transfers joint research activities Support to cooperation between firms Funding of S, T & I activities grants loans at preferential rate reimboursable advance (conditionned to success) loans garantee equity fundings/seed, risk capital export credits STI Policy rationales. Part I : Paradigms - L. BACH 2

Tax system research tax credit tax relief for technology-related purchase income tax on funds providers (business angels, foundations…) Legal and regulation aspects IPR Norms, technical reglementations Reglementation on foreign trade (techno transfer, barriers, …) Competences building (higher) education system Diffusion of scientific and technical information libraries, data base, info network Standard, plateforms, common langage Supporting infrastructures technical, legal, management …assistance and services scientific/technical facilities STI Policy rationales. Part I : Paradigms - L. BACH 3

Science - Technology - Innovation Policies : Long history, recent emphasis (WW II, 80/90 s) Many different tools and combination of tools (cf Georghiou-Edler tab) Fashion aspects, policy imitation and diffusion Multiple stakeholders Multiple decision levels A lot of typologies : Mission - diffusion, Vertical - horizontal, Supply side - Demand side, etc NEED OF THEORETICAL BACKGROUND : IDENTIFICATION OF RATIONALES, which to some extent are common to the "S", the "T" and the "I" dimensions of policy STI Policy rationales. Part I : Paradigms - L. BACH 5

Pecs Session / Week 2 - July 2009 STI policy rationales. Part I : paradigms • The basics : simple rationales for STI policy • The "traditional" opposition between paradigms • "Policy mix" and "rationale mix" STI Policy rationales. Part I : Paradigms - L. BACH 6

1. The basics : simple rationales for STI policy « Simple » rationales for State intervention in Science, Technology and Innovation (Pavitt - Walker) • Independance, security, prestige • Growth / industrial development through competitivity • Social development (education, health, jobs, Quality of Life, . . . ) • Scientific progress per se STI Policy rationales. Part I : Paradigms - L. BACH 7

Investment in STI, from innovator standpoint: • High (and increasing) R&D costs • Uncertainty on feasibility, results, economic value • imitation, knowledge leakage etc • demand reaction risk • long term rentability Lack of incentives for private R&D investment: = investment/profitabilty profile does not fit the "normal" private investment/profitability profile (NPV, payback, Ro. R…) Þ"too low" level of private investment as compared to the needs of society Þ PUBLIC INTERVENTION <=> simpliest formulation of "market failure" argument STI Policy rationales. Part I : Paradigms - L. BACH 8

Source : Philippe Bourgeois DGE/SPIC STI Policy rationales. Part I : Paradigms - L. BACH 9

2. The "traditional" opposition between paradigms • Detailed analysis of two paradigms Neo-classical / standard framework (NC) vs Evolutionist structuralist framework (ES) • Attempts to identify causal beliefs including : Theoretical foundations Main features « Failures » justifying State intervention Consequences of these failures Basic principles for State intervention (Tools / instruments) STI Policy rationales. Part I : Paradigms - L. BACH 10

Standard / Neo-classical framework : theoretical foundations - 1 Neo-classical / main stream economics + Theory of incentives : information asymetries + optimal contracts Transaction cost theory : governance cost + « at the fronteer » : (1) New growth theory (ROMER, LUCAS, AGHION, …) : Endogeneisation of S&T Importance of supply of knowledge (human capital, education, RD, infrastructure…) as a source of growth But focus on information, incentives, « mechanical » aspects => real departure from standard approach ? STI Policy rationales. Part I : Paradigms - L. BACH 11

Standard / Neo-classical framework : theoretical foundations - 2 (2) (New) Economics of science (DAVID, DASGUPTA, …) : • inherent and specific properties of the information : - Imperfect information (information paradox) - Non-rivalry and non-excludability (information as a "public good") => problem of property right - Low cost of REproduction • new line between S-related and T-related activities and outputs : - the practices of diffusion associated with incentive schemes - the choice of the optimal level of codification (cf reward system) - the higher uncertainty in the production and use of scientific results - the fact that results from basic research are considered mainly as a information input for applied research (more generic usefulness) - higher indivisibilities in science production - longer term perspective of science => possible background for distinction between S and T and I policies STI Policy rationales. Part I : Paradigms - L. BACH 12

"Linear model" invention BASIC RESEARCH innovation APPLIED RESEARCH DEVELOPPMENT diffusion "ON THE MARKET" Production Sales USE BY NONINNOVATORS Imitatosr, Adopters clients technology push science technology market demand pull • the "holy sequence" of clearly distinct steps • fit with classic distinctions : Nature of knowledge (public vs private), repartition of roles (State – univ - PROs vs firms), incentive mechanisms (Merton vs market) … STI Policy rationales. Part I : Paradigms - L. BACH 13

Standard / Neo-classical framework : Main features Comprehensive and extremely coherent « paradigm » • Market : unique mode of coordination and of selection • State is « outside » • Equilibrium • Static analysis • Optimizing rationality • Input - output perspective / linear model of innovation • Central focus : optimal allocation of resources • Normative reference : welfare/Pareto analysis • Research (S, T, I) as production of output = information + information as an input for downstream activities STI Policy rationales. Part I : Paradigms - L. BACH 14

Standard / Neo-classical framework : Knowledge (scientific, technological, productembedded etc) + production of knowledge + use of knowledge exhibit some characteristics not fitting with "ideal" characteristics => market/price mechanisms cannot work = market failures => consequences for social optimality => basic principles for public intervention STI Policy rationales. Part I : Paradigms - L. BACH 15

Standard / Neo-classical framework : failures, consequences principles for policy action • Imperfect information (information paradox) • Non-rivalry and nonexcludability => problem of property right • Low cost of REproduction • Indivisibilities, long term and • Lack of information on results, use and demand => high risk • Long term rentability • High cost • Problem of appropriability of S&T “products” and of gains from innovation => knowledge externalities => market externalities => network externalities Reducing uncertainty (environment, S, D) Substituting to the market (S and D sides) (sharing risk and cost) Allowing for internalizing externalities (property rights, cooperation) STI Policy rationales. Part I : Paradigms - L. BACH 16

Evolutionist structuralist framework : theoretical foundations Different approaches from different disciplines (mono or multi-disciplinary) with common features and specific focus Evolutionary theory : focus on evolution of technology, firms, industries, etc diversity generation / reproduction / selection processes Systemic / Network approaches (N/L SI, clusters, etc) : focus on coordination, complementarities, variety of institutions Knowledge-based economics : focus on knowledge creation, sharing, processing, access, diffusion, etc / cognitive processes STI Policy rationales. Part I : Paradigms - L. BACH 17

"interactive model" • More complex processus : feedback, overlaps • "Innovation" at all stages • Specific role of knowledge base: specific to each actor/organisation accumulation processes (capitalisation, learning, forgetfulness) => "path-dependancy" (example : Kline-Rosenberg model) • Classical distinction put in question • Linear model = one possible pattern within the interactive model • New perspectives on Nature of knowledge, repartition of roles, incentive mechanisms … STI Policy rationales. Part I : Paradigms - L. BACH 18

Evolutionist structuralist framework : main features NC framework • Market : unique mode of coordination and of selection • State is « outside » • Equilibrium • Static analysis • Optimizing rationality • Input - output perspective / linear model of innovation • Central focus : optimal allocation of resources • Normative reference : welfare/Pareto analysis • Research (S, T, I) as production of output = information + information as an input for downstream activities STI Policy rationales. Part I : Paradigms - L. BACH • Variety of modes of coordination and of selection • State is part of the game • No equilibrium • Dynamic analysis / Path dependancy • Other forms of rationality • Inter-active model of innovation • Central focus : creation of resources + knowledge (≠ information) = fundamental resource • Unclear normative reference : « adequate » system, processes, cognitive capacities ? environment ensuring « good trajectories » / « good paradigm » ? • Knowledge coming from anywhere in the system (not only Research) 19

Evolutionist structuralist framework : failures, consequences principles for policy action and • inadequate selection processes • lack of coordination, complementarities, lack of institutions, speed of adjustment between institutions and S&T. . . • misallocation of resources and cognitive attention between exploration and exploitation • knowledge creation, processing, distribution failures : codification, circulation, emitting/ absorptive / articulation capacity, structure of knowledge. . . • • lack of diversity “negative” lock-in difficulty for paradigmatic changes knowledge, social, institutional. . ”gaps” Not so coherent : Cognitive capacity of actors : development, orientation, adequate conditions of use. . . STI Policy rationales. Part I : Paradigms - L. BACH 20

NC framework ES framework Cognitive capacity of actors : development, Orientation, adequate conditions of use. . . Reducing uncertainty (environment, S, D) Substituting to the market (S and D sides) (sharing risk and cost) Allowing for internalization of externalities (property rights, cooperation) Tools / instruments Common ? Specific ? Optimal allocation by market (or pseudo-market) mechanisms Yes Social optimality Normative reference ? Diversity, selection, cohesion No State "a priori" in/out of system ? Out In STI Policy rationales. Part I : Paradigms - L. BACH « good » trajectories « good » transition between paradigms 21

Failures and rationales for policy action STI Policy rationales. Part I : Paradigms - L. BACH 22

STI tools re-interpreted STI Policy rationales. Part I : Paradigms - L. BACH 23

3. "policy mix" and "rationale mix" Beyond this NC / EC opposition : need to mix approaches ? Different rationales adressing different dimensions ? ex extented from A. BONACCORSI : (a) Incentives: neoclassical theory of market failure (b) Factors of production: endogenous growth (c) Processes and coordination: neoinstitutional and evolutionary (d) Learning: knowledge-based economics STI Policy rationales. Part I : Paradigms - L. BACH 24

Different (mix of ? ) rationales underlying one given tool/instrument § Tools which aim at fostering cooperation between actors : sharing/complementarity of costs, risks, information, joint creation of / distribution of knowledge ? § Public research : beyond « pure public good argument » ? § Patents : Protection/diffusion dilemna, signaling, intangible base for cooperation ? (see slides at end) STI Policy rationales. Part I : Paradigms - L. BACH 25

Mixing tools/instruments and mixing rationales ? • cooperative agreement /funding / IPR : NC : same failure adressed ? too many failure remedies ? (ex ante vs ex post reward vs patent ? ; limit fundings to transaction/cooperation costs ? . . . ) • policy oriented towards SME : Are supposed SME specificities grounded in rationales ? • Lisboa agenda and EC "policy mix": "the combination of policy instruments, which interact to infuence the quantity and quality of R&D investment in public and private sectors" (http: //www. policymix. eu/policymixtool/) • Research Infrastructures: sharing cost, create opportunity for networking, research outputs, research tools… (http: //cordis. europa. eu/infrastructures/) • The renewal of Demand side policy (cf Georghiou-Edler paper – next 3 slides) STI Policy rationales. Part I : Paradigms - L. BACH 26

The case of Demand side policy More on the "I" side Return of Demand Side policy in the policy arena (EU, UK), while still present in US and Japan Definition : All public measures to induce innovation and/or speed up diffusion of innnovations through increasing the demand for innovations, defining new functional requirement for products and services or better articulates demand STI Policy rationales. Part I : Paradigms - L. BACH 27

Mix of rationales : • Lack of incentives on demand side => fundings (NC) • Asymetries of information, transaction costs => increase and diffusion of info (NC) • Demand is often "local" and should be at least partly "locally" answered : ü heterogeneity, path dependancy, idosyncracy etc ü local Innov System, spillovers, etc ü user-producer interactions, "lead" user, etc => System and cognitive failures • "Good directions" towards orienting innovation processes => Selection failures • Demand Knowledge base should be aligned => System and cognitive failures (standards, etc) • Inter- gvtal department strategies and coordinations => System failures STI Policy rationales. Part I : Paradigms - L. BACH 28

Mix of STI policy tools: Basis = public procurement policies (general vs strategic, direct vs catalytic, commercial vs pre-commercial) Combined with supply side fundings - technology plateforms … Policy mixes : Improvement of public services and policies : better answers to societal need through innovation = sustainable devlopment, health, etc backed up by STI policy STI Policy rationales. Part I : Paradigms - L. BACH 29

policy boundaries, overlaps, complementarity : § ST&I vs competition policy (monopoly, cartels, public aids, public procurement…) § ST&I vs education policy (Univ. -PROs, longlife training…) § ST&I and environmental/sustainable dvpt oriented policy § ST&I and … STI Policy rationales. Part I : Paradigms - L. BACH 30

3. Towards a « rationale mix » framework for policy processes in reality (from EPOM / Prime No. E project http: //www. prime-noe. org/) Interaction between two types of rationales in the policy design, making and implementation processes Production policy rationales Governance policy rationales STI Policy rationales. Part I : Paradigms - L. BACH 31

Interaction between two types of rationales : 1. “Production policy rationales", i. e. causal beliefs, about the production of knowledge and instruments; providing a theoretical understanding knowledge creation and intervention (failure argument) and the type set-up of policy framework for justifying public of policy proposed <=> what was covered in first part of lecture Economics - Sociology of science "neo-classical paradigm" vs “evolutionist structuralist approach” (Lundvall & Borras 1997; Bach & Matt, 2005 …) STI Policy rationales. Part I : Paradigms - L. BACH 32

• Traditional paradigm (V. Bush) Focus on fundamental research, Justification of public research : defense, prestige, general welfare, knowledge per se + « pure public good » • Neo-classical / market paradigm (Arrow + Dasgupta-David) Distinction between fundamental research (open science, knowledge as public good) and technology (property right, Knowledge as private good) + competion / incentives / flexibility of resources / cost-cutting; Justification of public intervention : market failures • System / Network paradigm (Lundvall, Nelson, CSI…) Importance of the complementarity of /the links between multiple actors : coordination, alignement of objectives and resources…; Justification of public intervention : system failure • Evolutionary approach (Nelson & Winter, Dosi, Metcalfe…) focus on the generation (mutations), distribution (diversity) and diffusion (transmission) of changes + fitness and co-evolution as compared to the environment (selection mechanisms). diversity generation, diffusion • Knowledge-based approach (Cohendet Meyer-Krahmer) Knowledge as collectively produced, shared, distributed + multi-dimensional knowledge with tacit dimension + importance of learning processes; Justification of public intervention : learning (cognitive) failures STI Policy rationales. Part I : Paradigms - L. BACH 33

Strong connexions to sociology of science (Benner and Biegelbauer in EPOM [2005]) : • the simplest linear model, distinguishing Research and Economy, then enhanced by the Mertonian tradition; • standard linear model of innovation, with sequential (technology push or demand pull) link, and clear distribution of roles between actors along the steps of the model; • the interactive model stream, including Gibbons mode 2, Triple Helix and the like; • the constructivist approaches. STI Policy rationales. Part I : Paradigms - L. BACH 34

• Interaction between the actors – the Triple Helix (Loet Leydedorff Henry Etzkowitz) Researchers who are prepared to exploit their knowledge. Universities that profile their education and research. Research Entrepreneurs who recognize the value of new knowledge. Managers who can identify market requirements and have the courage to invest their resources. Investors who are prepared to wait for a return on their investments. Politics Business Legislators who influence the conditions under which companies and researchers are working. Politicians who allocate resources for research and development. Local authorities and county councils who set out to facilitate the establishment of new companies and make their local areas pleasant places to live. Adapted from M. Benner [2005] Separate institutions / intermediaries - Modes of direct communications - Mixed roles STI Policy rationales. Part I : Paradigms - L. BACH 35

Interaction between two types of rationales : 2. “Governance policy rationales”, reflecting the governance paradigms ruling state intervention in general; not policy sector specific but have encompassing validity; they often correspond to political traditions and culture Political science - Public management STI Policy rationales. Part I : Paradigms - L. BACH 36

• Centralism / technocratic model Centralization of decision processes Highly professionalized civil servant Command & Control model strong hypothesis on the capacity of State in terms of access to information, processing of information and action • Network State model with a focus on coordination role, decentralisation, enablement skills, public/private cooperation, self-regulatory approach • New Public Management model Clear policy target/goals Clear budget Systematic performance analysis Clear and explicit decision processes • Decentralized multi-level model : multiple centers of decision with budget, staff…(not necessarily hierarchical) • Decentralized multi-space model : multiple and heterogeneous public & scientific interest groups (public opinion, consumers, patients, NGO, …) STI Policy rationales. Part I : Paradigms - L. BACH 37

Governance Policy Rationales Production Rationales Policydesign & making Incl. Monitoring and evaluation tools (+ "no instrument") Policy tools Policy Design & making frame Implementation STI Policy rationales. Part I : Paradigms - L. BACH 38

Policy design/making frame • Public influence and the business point of view • "Fertile soil", i. e. background and receptivity of policy makers (political preferences, culture, education, …) => selective attention and cognitive choices of policy makers • Turnover of political personal and of technocrats • Policy entrepreneurs • Boundary institutions • Examples, images and stylised facts ("prototype embodying knowledge") • Reports and other white papers • "Tactical interests" • Policy-making procedure and its "hidden side" • Competing rationales in government STI Policy rationales. Part I : Paradigms - L. BACH 39

Not static / sequential / "once and for all" Dynamics : • Path-dependancy • Learning and feed-back loops • Continuous (minor) changes • Major revisions triggered by : üinefficiency observed in the system ümajor changes in ideology or rationales üdiffusion of ideas ( « emulation » , benchmarking) üpressure related to external shocks or the public => time matching between policy cycle / ideas cycle (windows of opportunity STI Policy rationales. Part I : Paradigms - L. BACH 40

mix Production Rationales mix dynamic coherence co-evolution Policydesign & making Governance Policy Rationales mix Policy tools mix Implementation mix STI Policy rationales. Part I : Paradigms - L. BACH 41

Towards « archetypal » mixes ? 3 coherent models Production Rationale ideal model 1 Traditional Research & Innovation modes Simpliest linear model Research / Economy ideal model 2 Neo-classical / market System / Network Evolutionism Knowledge-based Governance Rationale Centralism / technocratic Command&Control Merton Standard linear model Interactive model New Public Management Network State / Gibbons mode 2 Triple Helix Constructivism STI Policy rationales. Part I : Paradigms - L. BACH Decentralized multi-level Decentralized multi-space 42

The case of patent : NC vs K. O. framework (based on [COHENDET-MEYER-KRAHMER, 2004 - PENIN, 2004]) Patent in NC revisited approach : n Knowledge as information + appropriability pb § Patents = appropriability means, strong property right=> to enhance individual motives (individual inventor), BUT should favour global knowledge production + appropriability / diffusion (+ cumulativeness) dilemna => Various length and scope of patent STI Policy rationales. Part I : Paradigms - L. BACH 43

Patent in Knowedge-based approach (1) : n Patents = appropriability + other dimensions: o negotiation role (bargaining power) o first step for cooperation/knowledge exchange (balance of power between members of network) o Signalling/disclosure device/reputation Example: communities (free-software) produce semi-public goods (common but not available to all) STI Policy rationales. Part I : Paradigms - L. BACH 44

Patent in Knowedge-based approach (2) : Reconsidering incentives (ex. free software) : § the key role of institutional settings (IS=norms, rules, standard) § IS govern incentives to produce and diffuse knowledge § IS shape the codification processes and thus the costs of transfering knowledge (science vs. industry) => If incentives to build knowledge in a community are strong, then appropriation is marginal STI Policy rationales. Part I : Paradigms - L. BACH 45

Patent in Knowedge-based approach (3) : § Is patent still a valid policy tool ? new dimensions of patents <=> new uses made by firms => new «K. O. failures» ? does patent help to overcome those «K. O. failures » ? If yes, is it an appropriate tool? (ex. : too expensive as a tool to foster cooperation!) § Patents hamper diffusion (traditional view), BUT also the production of knowledge (ex: IPR on software; IPR on fragments of gene before identification of product => no product => go beyond cumulativeness of information : importance of common cognitive platforms). STI Policy rationales. Part I : Paradigms - L. BACH 46