Firms’ research and development (R&D) efforts are known to generate spillover effects on other firms’ outcomes, e.g., innovation and productivity. Policy recommendations that ignore spillover effects may not be optimal from a social perspective whence the importance of accounting for spillover effects. Quantifying R&D spillover effects typically requires a spatial matrix that characterises the structure of interaction between firms. In practice, the spatial matrix is often unknown due to factors that include multiplicity of forms of connectivity and unclear guidance from economic theory. Estimates can be biased if the spatial matrix is misspecified, and they can also be sensitive to the choice of spatial matrix. This dissertation develops robust techniques that estimate the spatial matrix alongside other parameters from data using a two-pronged approach: (1) model elements of the spatial matrix using spatial covariates (e.g., geographic and product market proximity) and a parameter vector of finite length and (2) estimate the spatial matrix as a set of parameters from panel data. Approaches (1) and (2) address two identification challenges - uncertainty over relevant forms of connectivity and high-dimensionality of the design matrix - in single-index models. In this three-chapter dissertation, the first approach is applied in the first and third chapters, while the second approach is applied in the third chapter. Chapter 1 proposes a parsimonious approach to estimating the spatial matrix and parameters from panel data when the spatial matrix is partly or fully unknown. By controlling for several forms of connectivity between firms, the approach is made robust to misspecification of the spatial matrix. Also, the flexibility of the approach allows data to determine the degrees of sparsity and asymmetry of the spatial matrix. The chapter establishes consistency and asymptotic normality of the MLE under conditional independence and conditional strong-mixing assumptions on the outcome variable. The empirical results confirm positive spillover and private effects of R&D on firm innovation. There is evidence of time-variation and asymmetry in the interaction structure between firms. Geographic proximity and product market proximity are confirmed as relevant forms of connectivity between firms. Moreover, connectivity between firms is not limited to often-assumed notions of proximity; it is also linked to past R&D and patenting behaviour of firms. Single-index models suffer non-identification due to rank deficiency when the design matrix is high-dimensional. Chapter 2 proposes an estimator that projects a high-dimensional parameter vector into a reduced consistently estimable one. This estimator generalises the assumption of sparsity which is required for shrinkage methods such as the Lasso, and it applies even if the high-dimensional parameter vector’s support is bounded away from zero. Monte Carlo simulations demonstrate high approximating ability, improved precision, and reduced bias of the estimator. The estimator is used to estimate the network structure between firms in order to quantify the spillover effects of R&D on productivity using panel data. The empirical results show that firms on average generate positive R&D spillovers on firm productivity. The spatial autoregressive (SAR) model has wide applicability in economics and social networks. It is used to estimate, for example, equilibrium and peer effects models. The SAR model, like other spatial econometric models, is not immune to challenges associated with misspecification or uncertainty over the spatial matrix. Chapter 3 applies the approach developed in Chapter 1 to estimate the spatial matrix in the SAR model with autoregressive disturbances in a parsimonious yet flexible way using GMM. The asymptotic properties of the GMM estimator are established, and Monte Carlo simulations show good small sample performance. / Economics

The purpose of this thesis is to investigate the impact of government non-investment and applied R&D subsidies on the firm level competitiveness. According to many literature sources the indicators of competitiveness are assumed to be Return on Assets, Return on Equity, Return on Sales and Asset Turnover ratio. The tool used to test the effect is counterfactual analysis. The subsidized companies are compared to a control group drawn by Propensity Score Matching, the effect is tested by Average Treatment Effect on the Treated and the difference in size of other financials is derived by Difference in Differences. The study focuses on two industries NACE C25 (metallurgy) and C28 (manufacture of machines). The subsidized companies show no added significant growth of competitiveness whatsoever. The financials (Total Assets, Equity, Turnover) grew, compared to the control group, only for industry C25. JEL Classification O31, O38, H21 Keywords R&D, Innovation, non-investment, subsidies, applied R&D, competitiveness Author's e-mail petr.rotr@post.cz Supervisor's e-mail petr.teply@fsv.cuni.cz

Este trabalho estuda o processo de composição da carteira de projetos de P&D em casos da industria brasileira de produtos tangíveis. Com esse propósito, fez-se uma revisão bibliográfica e, em seguida, um estudo exploratório de múltiplos casos do tipo holístico, para identificar os procedimentos, critérios e participantes envolvidos no processo, assim como as dificuldades enfrentadas. O estudo apresenta um modelo de referência de pesquisa elaborado a partir das contribuições da teoria, envolvendo as diferentes abordagens sobre o processo de composição da carteira. O modelo serve também de marco de referência para a coleta, classificação, análise e comparação dos dados, e proporciona a flexibilidade necessária para o desenvolvimento de estudos deste tipo. Os resultados da pesquisa mostram que as carteiras são resultantes de um processo no qual a avaliação de projetos no contexto de carteira é, na sua maior parte, não formalizada ou inexistente. Adicionalmente, os resultados mostram que o processo envolve participantes dos vários níveis hierárquicos das empresas, os quais contribuem direta ou indiretamente com as decisões do processo de composição da carteira. Finalmente, os resultados evidenciam que a obtenção de informação confiável e completa, a distribuição de recursos e a burocracia são as principais dificuldades enfrentadas ao longo do processo. / This work studies the R&D project portfolio composition process in cases of the Brazilian tangible products industry. For this purpose, a bibliographic review was done, and then, an holistic multi-case exploratory study, to identify the procedures, criteria and participants involved in the process, as well as the difficulties faced. A study presents a research reference model, constructed from the theory contributions, and involves several approaches of the portfolio composition process. The model is also useful as a framework for data collection, classification, analysis and comparison, and provides the flexibility needed for developing this kind of research. The study shows that the project portfolios result from a process where the project portfolio assessment is mostly informal or doesn?t exist. Additionally, the study shows that the process involves participants of several hierarchical levels in companies, where each participant contributes in a direct or indirect way, in decisions of the portfolio composition process decisions. Finally, the results identify that reliable and complete data collection, resources distribution and bureaucracy are the main difficulties faced along the process.

Face aux enjeux environnementaux actuels, s’investir dans les énergies renouvelables au nom de la « transition énergétique » est l’alternative la plus adoptée par de nombreux pays. Les énergies renouvelables, en particulier l’énergie photovoltaïque, sont ainsi devenues des secteurs concurrentiels innovants en pleine expansion. Dans ce contexte, il est donc nécessaire de s'assurer que les nouvelles technologies PV, qui sont complexes et issues de plusieurs étapes de fabrication, répondent aux critères d'un produit peu impactant pour l’environnement, dénommé ici éco-conçu. L’état de l’art sur l’éco-conception montre que la prise en compte des contraintes environnementales dans les projets de R&D à échelle TRL « Technology Readiness Level » bas est encore un phénomène émergent, pour les raisons suivantes : d’une part l’évaluation de l’impact environnemental est relativement complexe pour une technologie non mature en cours de développement (TRL bas) ; ses caractéristiques et procédés de fabrication n’étant pas encore tous connus ; d’autre part les outils identifiés présentent plusieurs écueils qui limitent leur intégration dans les organismes PV de R&D.Ce travail de recherche vise donc à développer une méthodologie pour permettre l’intégration pérenne de l’éco-conception dans les organismes de R&D, pour accompagner leurs partenaires industriels dans leur effort d’innovation et de compétitivité.La méthode développée est ainsi basée sur le principe d’estimation des taux d’évolutions techniques, économiques et environnementales d’une nouvelle technologie (TRL bas) par l’intermédiaire d’une base de données de référence. La construction de cette dernière s’appuie sur l’Analyse du Cycle de Vie (ACV) comme outil de pilotage pour fournir des résultats fiables, malgré le faible niveau TRL.Pour intégrer de façon pérenne cette approche au sein des organismes de R&D, la méthode développée a été implémentée dans un outil informatique « ECO PV » dédié aux systèmes PV en silicium cristallin, qui constituent plus de 90% du marché PV actuel. Basé sur le principe de démocratisation des connaissances environnementales et de la capitalisation des informations, cet outil est accessible non seulement aux experts en ACV mais aussi à l’ensemble des ingénieurs du domaine PV.Ce travail de recherche a enfin permis d’une part de générer des premiers résultats à la fois fiables, simples et quantifiés, et de développer une méthodologie d’éco-conception pour orienter les choix technologiques des projets dans les phases amont de R&D, afin de développer des systèmes PV plus respectueux de l’environnement. / Faced with current environmental challenges, investing in renewable energies in the name of the "energy transition" is the alternative most adopted by many countries. Renewable energies, in particular photovoltaic solar (PV) energy, have thus become innovative and competitive sectors in full expansion. In this context, it is therefore necessary to ensure that new PV technologies, which are complex and the result of several manufacturing stages, meet the criteria of a product with low environmental impact, referred here as eco-designed.The state of the art on eco-design shows that the consideration of environmental constraints in R&D projects with low TRL "Technology Readiness Level" is still an emerging phenomenon, for two main reasons. On the one hand, environmental impact assessment is relatively complex for a non-mature technology under development (TRL low) because its characteristics and manufacturing processes are not yet fully defined. On the other hand, the identified tools in the literature have several limitations that impede their appropriation in R&D organizations.This research work therefore aims to develop a methodology to enable the sustainable integration of eco-design into R&D organizations, supporting their industrial partners in innovation and competitiveness. Thus, the developed method is based on the estimation of the evolution rate of technical, economic and environmental criteria of a new technology (low TRL) through a specific database of reference. The construction of the database relies on Life Cycle Assessment (LCA), used as a management tool to provide reliable results, despite the low TRL level.To integrate this approach into R&D organizations in a sustainable way, the method developed was implemented in a software "ECO PV" dedicated to crystalline silicon PV systems, which represent more than 94% of the current PV market today. Based on the principle of democratization of environmental knowledge and capitalization of information, this tool is accessible not only to LCA experts but also to all engineers in the PV field.Finally, this research work enabled to generate reliable, simple and quantified results and to develop an eco-design methodology to guide the technological choices of projects in the upstream phases of R&D, in order to develop PV systems more environmentally friendly.

Corporate taxation is a very politicized topic, and policymakers have different perspectives on what level of corporate tax rate yields the optimal outcome for research and innovation activity. The argument is divided where one side believes that corporate taxation and R&D activity has a negative relationship, which implies that a decrease in corporate taxation yields a better outcome for firms’ R&D activity. Whilst the other side believes that there exists a positive relationship, implying that the prevailing strategy is to increase corporate taxes to further encourage R&D activity. Therefore, the inherent purpose of this study focuses on the effect that corporate taxation has on the U.S corporate environment, whilst researching to understand the two polarized perspectives on corporate taxation and discover which perspective appears more prominent. To succeed in this regard, this study extracts internal financial information from firms located in the U.S states, by comparing the corporate tax change that the Trump administration implemented in 2018. Enabling this research study to compare firms’ R&D expenditure during the pre-tax reform (2017) with the post-tax reform (2018), indicating a firm’s differences in R&D expenditure. The difference-in-difference method was used with a total sample size of 71 firms, which uses a control group and a treatment group to negate macro-environmental disruptions to indicate the accurate effects of the tax policy change. The results state that companies that experienced a reduction in corporate taxes raised their R&D expenditure, approximately by 11.60%. This showed that there was a very large increase in R&D activity when corporate taxation was significantly lowered, which would support the literature’s findings that suggested the existence of a negative relationship. However, the research yielded a result that would indicate that the 11.60% increase was not a statistically significant result and that further studies need to be conducted.

Les startups innovantes orientées technologies mènent de nombreusesexpérimentations pour recueillir suffisamment de preuves de valeur, et réduire ainsiles incertitudes avant le lancement de leurs produits et services sur le marché. Bienque les moyens des startups soient limités, ces expérimentations sont souventaléatoires, itératives et nécessitent un investissement important en termes de tempset de ressources. Des solutions doivent ainsi leur être apportées afin de les aider àmieux planifier la maturation de leur technologie et à mieux identifier les opportunitéssur le marché. C’est dans ce contexte que s’inscrit la présente recherche.Une revue de la littérature est d’abord menée dans les domaines de l'ingénierie de laconception ainsi que du management de la technologie et de l'innovation. Cetteanalyse démontre que les méthodes et outils existants ne permettent pas d’établir desfeuilles de route précises et quantitatives pour mener des expérimentationspertinentes. Une enquête est ensuite réalisée sur les pratiques de 60 startupsinnovantes dans le contexte de la Ville de Paris. Cette recherche révèle que malgrél’utilisation des méthodes d’innovation telles que Lean Startup et BMC (BusinessModel Canvas), les startups n’ont aucun outil pratique pour spécifier et planifier leursexpérimentations, ou pour allouer efficacement leurs ressources.Par conséquent, cette recherche apporte trois contributions. D'abord, l'outil BMC estcombiné avec la quantification des poches de valeur (opportunités) et une séried’indicateurs quantitatifs permettant d’assurer la pertinente de l'offre de valeur.Ensuite, une méthodologie dénommée DUE (Design by Usage-based Experimentation)est proposée afin d’examiner l'ensemble des marchés potentiels et d’identifier lesprincipales propriétés technologiques à améliorer grâce à l’expérimentation. Uneméthodologie quantitative nommée RITHM (Roadmapping Investments in TecHnologyand Market) est par la suite proposée pour optimiser les expérimentations pourqu’elles puissent aboutir à la maturation d'une technologie pour atteindre les marchésles plus profitables. La pertinence de ces méthodologies est évaluée en les appliquantà une startup orientée technologies et par la validation d’experts. Enfin, cette rechercheconclue qu'il est possible d’améliorer la spécification des feuilles de routed’expérimentations dans une startup technologique et de mieux orienter sesinvestissements vers des stratégies R&D plus efficaces et économiques.Cette recherche peut contribuer à soutenir de manière significative la prise de décisionde différents acteurs, comme par exemple, les entrepreneurs, qui doivent justifier desdépenses de R&D dans des dossiers de levée de fonds, ou des investisseurs publicset privés qui cherchent à minimiser les risques d'investissement dans destechnologies innovantes, ou enfin des technology scouts. / Technology-oriented innovative startups must conduct extensive experiments to gather sufficient proofs of value, so that they may reduce uncertainties before the launch of their products and services on the market. Despite their limited resources, the experiments conducted by startups are often random, loop-based and consequently time and resource consuming. Hence, startups need solutions to help them better planning their technology maturation and identifying market opportunities. It is in the context that the current research is carried out.First, a literature review is conducted in the fields of engineering design, technology management and innovation management. This review suggests that the existing methods and tools fail to provide accurate and quantitative roadmaps to conduct relevant experiments. Second, this research surveys the design and testing practices of 60 innovative startups in the context of the City of Paris. This investigation reveals that despite the use of popular innovation methods such as Business Model Canvas (BMC) and Lean Startup, startups have no practical tools for specifying and planning experiments, nor for efficiently using their resources.Therefore, this research makes the following three contributions. First, the BMC tool is combined with quantified value buckets (opportunities) and innovativeness indicators in order to make sure that the value offer is effective and significant. Second, a Design by Usage-based Experimentation (DUE) methodology is proposed to help screening the whole set of potential markets and to identify the main technology properties that need to be improved through experiments. Third, a quantitative methodology called RITHM (Roadmapping Investments in TecHnology and Market) is proposed to optimize the set of experiments, so that they result in the maturation of a technology to reach the most profitable markets. The relevance of the proposed methodologies is assessed through their application to technology-oriented innovative startups and also through expert validation. Finally, this research concludes that it is possible, with more effective and economical R&D strategies, to better define experiment roadmaps and better steer technological startup investment.This research may significantly support the decision-making process of various actors: entrepreneurs, who need to justify R&D expenses in fund-raising applications; public and private investors, who constantly tend to minimize investment risks in innovative technologies; or technology scouts.

Par la promotion de l’innovation en réseau, les instruments de dynamiques territoriales sont devenus aujourd’hui des outils de support indispensables à la compétitivité des entreprises et à la croissance des nations. C’est dans cet esprit que sont conçus les pôles de compétitivité dans le dispositif français de 2005. L’innovation collaborative à l’échelle des pôles de compétitivité vise en premier lieu à créer des débouchés nouveaux, c’est-à-dire à générer des innovations prêtes à être diffusées sur les marchés. Or, si les réseaux d’innovation ont été largement étudiés à travers des outils de pilotage, de gouvernance et de coordination, par la finance, la propriété intellectuelle ou les pratiques de RH, par la formation, par la confiance entre les partenaires, par les externalités, ou encore par la dynamique des PME, une carence persiste dans les recherches actuelles à propos de l’orientation stratégique composite des réseaux d’innovation, en particulier l’orientation marché ou client dans la dynamique des projets/acteurs. Les premiers rapports d’évaluation des Pôles dénotent au demeurant, des écueils relatifs à une collaboration propice à des innovations prêtes pour l’accès aux marché et la stabilisation de problèmes liés au fonctionnement, à la gouvernance et à la performance des pôles. Tout en nous inscrivant dans le courant des travaux de recherche qui mettent en exergue les spécificités du dispositif français des pôles de compétitivité composé d’acteurs complémentaires (Recherche, Formation et Entreprises), notre travail met l’accent sur le fait que l’innovation collaborative est mise en tension par des intentions stratégiques différenciées, liées précisément aux profils hétérogènes, et parfois divergents, des partenaires animés par un projet commun. Par une analyse empirique portant sur le pôle SCS que nous avons choisi en raison de son profil, de la nature de ses projets et de son historique collaboratif avec notre laboratoire de recherche (abondance de données préexistantes), nous tentons de comprendre la façon dont s’intègre l’orientation « marché » dans les projets d’innovation en réseau, et le rôle que la dimension marketing joue dans la dynamique d’innovation collaborative au sein des pôles de compétitivité. L’étude « qualimétrique » conduite a permis de mettre en évidence une relation de dépendance réciproque très forte entre les particularités des déterminants de l’innovation collaborative, notamment les profils d’acteurs et les intentions stratégiques, et les modalités d’intégration de l’orientation client/marché dans les dans les projets d’innovation collaboratifs (PIC). L’attribut « réseaux de connaissances » qui distingue les PdC octroie au marketing une place limitée dans les PIC. Aujourd’hui, une révision du modèle assimilable à un « réseau connaissances » s’impose, car les résultats montrent aussi une limite des marges de manœuvre des Pôles qui peut constituer une entrave à l’accès aux marchés des innovations issues des PIC. En effet, le processus d’innovation collaborative à l’échelle des PdC ne permet qu’au plus de parvenir à dresser des prototypes (β Product) et n’octroie pas de possibilité d’aller au-delà (test, production, mise sur le marché, etc.). Nos contributions théoriques et managériales permettent aujourd’hui d’ouvrir des grilles d’analyses nouvelles et d’améliorer le fonctionnement, la gouvernance et la performance des pôles à travers le succès des projets d’innovations collaboratifs. La thèse nous a aussi amené à développer des hypothèses nouvelles qui pourraient être échafaudées dans le cadre de recherches ultérieures, notamment au regard de la pertinence des réseaux de connaissances dans le processus complet d'innovation collaborative. / By promoting networked innovation, instruments of territorial dynamics have now turn out to be essential tools supporting firms' competitiveness and growth of nations. Clusters of competitiveness (CC) designed by French government in 2005 follows that state of mind. The first goal of collaborative innovation within CC is to create new opportunities and generate innovations ready to be diffused on new markets. However, if innovation networks have been widely studied through many questions (such as monitoring tools, governance and coordination, with finance, intellectual property or HR practices, training, trust between partners, externalities, or by the dynamics of SMEs), a deficiency persists in current researches about the strategic orientation of the composite innovation networks, more particularly with the customer or market orientation in the dynamics of projects / actors. Moreover, the first evaluation reports of the system had put a focus on pitfalls considering its capacity to generate innovations ready for market access and the stabilization of operation difficulties, as well as some governance and performance problems. While placing our analysis within current researches which highlight the specificities of the French clusters of competitiveness composed of complementary protagonists (Research, Academic and Enterprise), we intend to focus on the fact that collaborative innovation is tensioned by differentiated strategic intentions and profiles specifically related to heterogeneous, and sometimes conflicting, partners driven by a common project. By an empirical analysis carried out on the SCS French cluster chosen owing to its profile, the nature of its projects and the background history of collaboration with our laboratory (a wealth of existing data); we try to understand how the customer orientation fits into the clusters, and the role that marketing plays in the dynamic collaborative innovation within clusters of competitiveness. Our “qualimetric” study has allowed us to bring out a very strong mutual relationship of dependency between the particularity of collaborative innovation determinants, notably profiles of actors and strategic intentions, and how to integrate the customer / market orientation within the collaborative innovation projects. As "knowledge networks", CC grant to marketing and customer orientation a limited role in the collaborative innovation projects. Today, a revision of the model similar to a "knowledge network" is needed because our results seem also to reveal a limited room for maneuver of clusters of competitiveness, which can be for the collaborative innovation projects an obstacle for generating innovations ready for market access. Indeed, in its current state the process of collaborative innovation on a clusters of competitiveness scale enable only to obtain prototypes (β Product), and does not permit to go beyond (ie. test, production, market, etc.). Our theoretical and managerial contributions allow to open up new analytical tools as well as to improve the operation, the governance and the performance of cluster competitiveness through the success of collaborative innovation projects. The thesis has also led us to develop new hypotheses, which could be built up through further researches, in particular with regard to the relevance of knowledge networks in the whole process of collaborative innovation.

The Thesis is composed by three complementary research investigations on the economic and policy aspects of EU corporate R&D.Collectively, the work first reviews the theoretical and empirical literature of corporate R&D intensity decomposition; it then investigates the EU R&D intensity and its decomposition elements comparatively with most closed competitors and with emerging economies over the period 2005-2013. Finally, it inspects further some key aspects that can be associated to the EU R&D intensity gap: sectoral dynamics and the resulting sectoral and technological specialisations as well as the drivers for R&D investment growth across sectors and firms' age groups of top R&D investing firms over time. These studies also address the possible policy implications that derive from their outcomes.The investigations rely on literature as well as on company data, mainly from nine editions (2006-2014) of the EU Industrial R&D Investment Scoreboard. For analytical purposes they use literature review, meta-analysis, descriptive statistics, R&D intensity decomposition computational approach, Manhattan distance and Technological Revealed Comparative Advantage metrics, and a multinominal logit regression model. The results of these three research works are novel in several aspects. It indicates that literature results on R&D intensity decomposition differ because of data and methodological heterogeneities, and that the structural cause is the main determinant of EU R&D intensity gap if sector compositions of the countries are considered. It inspects how the use of different data sources and analytical methods impact differently on R&D intensity decomposition results, and what the analytical and policy implications are.The empirical research results of this Thesis confirm the structural nature of the EU R&D intensity gap. In the last decade the gap between the EU and the USA has widened, whereas the EU gap with Japan has remained relatively stable. In contrast, the emerging countries' R&D intensity gap compared to the EU has remained relatively stable, while companies from emerging economies are considerably reducing such gap. Besides, as novel contribution to the state of the art of the literature, this Thesis uncovers the differences between EU and US by inspecting which sectors, countries and firms are more accountable for the aggregate R&D intensity performance of these two economies, and it finds a high heterogeneity of firms' R&D intensity within sectors. Furthermore, it shows that there is a bigger population of both larger and smaller US top R&D firms which invest more strongly in R&D than competitors, and that the global R&D investment is concentrated in a few firms, countries and industries. Finally, the research founds a slightly higher EU R&D shift over sectors compared to the US, but not strongly enough towards high-tech sectors. Also, the EU has an even broader technological specialisation than its already broad industrial R&D sector specialisation, while the USA leads by number of technological fields belonging mostly to the industrial R&D sectors of its specialisation. Furthermore, the EU has been better able than the USA and Japan to maintain its world share of R&D investment even during the years of economic and financial crisis. Lastly, the study also indicates that firms make a complementary use of capital expenditures and R&D intensity for their R&D investment growth strategies and it reveals that there are differences in their use between firms' age classes across sectors. Overall, the main results of the Thesis suggest that to reach a more positive R&D dynamics and boost its competitiveness, the EU should adapt its industrial structure and increase the weight of high R&D intensive sectors. A focus on creating the conditions for firm creation and growth in new-emerging innovative sectors is advised together with favouring the exploitation of the full capacity of EU leading - but mature - sectors to also absorb high-technology from other sectors. / Doctorat en Sciences économiques et de gestion / info:eu-repo/semantics/nonPublished

由於資訊科技的發達，使全球資訊的分享加速，一方面使得企業創新的刺激源源不絕，另一方面也使得消費者的需求慾望越來越廣，造成產品生命週期不斷縮短，在現今的環境下，企業若要長久經營甚至維持成長，就必需積極從事創新之新產品開發。雖說新產品或技術的創新對企業的貢獻相當大，但研發專案往往存在高度不確定性且耗時耗力，一旦專案失敗，亦會對企業造成相當程度的傷害，導致許多企業望之卻步。因此，1960年代初期，Mansfield(1968)等學者開始從事研發支出費用與產業或公司生產力相關性的研究。其後，由於財報資料取得的方便性與客觀性，加上個別研發專案的效益較難追蹤、亦難以評估，大多數研發投入與研發績效相關的研究，首先，在變數上都是以財報上的財務數字為基礎，如研發費用、研發密度、公司獲利、營收等；其次，在研究個體上則是以企業整體為主。然而，過去許多研發投入與績效相關之研究，不論在投入面或效益面皆提出了許多非財務性的項目，此外，在現今即將邁入第四代研發管理的環境下，組織企業的研究發展工作多以「專案」的方式來進行，而非以年度企業整體預算進行管理。綜合上述兩點過去此類研究與實務上之落差，本研究採用經濟部工業局「主導性新產品開發計劃」之計劃成效追蹤問卷資料，以個別研發專案為研究對象，並將投入面擴大區分為成本、人力、時間三個構面，探討研發專案投入對非財務面之技術產出及財務績效之影響，期能彌補過去此類研究與實務上之落差。 本研究依照研發專案投入所產生之非財務與財務績效構面，分成兩大研究主題，研究結果顯示： 一、研發專案投入對技術產出之影響：專案投資金額對專利權與創新技術數有顯著正向的影響，專案研發人員之教育程度只對創新技術數有顯著正向之影響，研發時間與專案技術產出皆無顯著相關。 二、研發專案投入對財務產出之影響及其遞延效果：專案投資金額與專案研發人員之教育程度對開發產品銷售額具有顯著之正向影響，且從產品銷售第一年起，沒有時間上的落差，且專案投資金額之影響可延續三年，專案研發人員之教育程度之影響可延續兩年，研發時間與開發產品銷售額無顯著相關。 此外，吾人並針對實證結果與實務界人士進行訪談，瞭解與假說不符者之原因，結果發現：專案研發時間較長，不完全是因為企業願意投入較多的心力，有很大的可能是研發瓶頸的發生與研發人員的不效率，這些部份不但對專案產出沒有顯著貢獻，還可能危害專案的成功。 / As a result of the progression of information technology, the speed of information sharing is getting higher and higher. On one hand it makes companies get the stimulation of innovation more easily than before, on the other it also makes the desire of customers becomes wider and wider. Thus, the life cycles of all products in the world then become shorter and shorter. In this environment, companies who want to survive or even get growth have to do their development and research activities more aggressively than before. Although the benefits of R&D is great, lots of companies still shrink back at the sight. That’s because of the high uncertainty of R&D and the needs of huge amount of time and money, and companies always get incredible harm once they fail. Therefore, many researchers have tried to clarify the relationship between R&D inputs and R&D performance. Due to the accessibility and objectivity of financial report, the past researches in this field always used the information from financial reports, and set their research objects as the whole companies. But many researchers have pointed out that there are not only financial inputs and outputs of R&D activities but also non-financial ones. In the global trend to 4th generation of R&D management, R&D organizations have changed their management way form total budgeting to project management. To eliminate these two gaps between theory and practice, in this research we use the questionnaire data of Leading Product Development Programs from Industrial Development Bureau Ministry of Economic Affairs to make our research stand on a project view point and get other non-financial inputs and outputs data that we can’t get from financial reports. Hope we can earn more practical results to help companies making their decisions of R&D activities. According to the two perspectives of R&D project performances, non-financial and financial, we separated this research into two subjects, the results showed that: 1. The effects of R&D project inputs on technical outputs: we found that the project dollar investment has significant effect on the number of patents and innovative techniques, and the education degree of the project’s R&D engineers only has significant effect on innovative techniques but on patents. On the other hand, R&D time of the project has no significant effect on both patents and innovative techniques. 2. The effects of R&D project inputs on financial performance, their time-lag effect and their continuity: we found that project dollar investment has significant effect on sales of the product that developed from the project since the first year it was sold, which means the time-lag doesn’t exist in this situation, and this effect can last for three years or even more. The education degree of the project’s R&D engineers has significant effect on sales of the product that developed from the project, from the first year it was sold. So, there is no time-lag, but this effect can only last two years. Besides, R&D time of the project has no significant effect on any year’s sales of the product that developed from the project. Finally, we have interviewed some practitioners discussing about our results that were not consistent with our hypotheses. According to their opinions, the reason why some projects spend longer R&D time is not just because those companies intend to pay more effort on those projects. It also includes the inefficiency of R&D engineer and the bottleneck of that project. These inefficiency and bottlenecks are not only non-value-added but also sometimes harmful to a R&D project. That’s why we found in this research that there is no significant relationship between R&D time of a project and its performances, neither financial performances nor non-financial ones.

No description available.