INNOVATION ET BREVETS
Econométrie de la Propriété industrielle

Grand Amphithéâtre Lumière - Lyon
14 & 15 Mai 1998

Résumés des interventions / Abstracts

Innovamétrie
PATENTS, HOME MADE DATABASES AND AUTOMATIC BENCHMARKING

H. Dou
University of Aix-Marseille - CRRM
Scientific Center of Saint Jérôme
13397 Marseille Cedex 20

M. Crener
U.S.E. University  of Southern Europe
2 rue Prince Héréditaire Albert
98000 Monaco

J.-M. Dou
I.D.S. Europôle Méditerranéen de l'Arbois
Domaine du petit Arbois
BP 67 - Les Milles
13762 Aix en Provence Cedex

New facilities to the access of Patent information with low cost expenditure can be achieve easily to day. This reason prompt us to present the various steps which can induce various companies, even SME, to built up in house, information systems able to give a better view of competitors and of the technology available. These think tanks can be built up around various products which will develop an innovation thinking of the company through comparisons techniques, graphs, networks and automatic benchmarking. This paper will deal particularly with bench marking techniques.

TECHNICAL STANDARDS, INDICATORS FOR THE DIFFUSION OF INNOVATIONS, AND THEIR RELATIONSHIP TO PATENT STATISTICS: THE GERMAN EXPERIENCE

K. Blind
Fraunhofer Institute for Systems and Innovation Research
Breslauer Str. 48
D-76139 Karlsruhe (Deutchland)

Introduction : In times of higher dynamics in technical change, the role of technology diffusion gains in importance. Technical standards are a visible indicator for the diffusion of new technologies. This leads to the focus on the statistical relationship between innovations measured by patents and standardisation measured by so called de jure standards edited by standard development organisations (SDOs). One causal connection is based on the following idea : A pool of innovations will be condensed into a set of technical standards. Therefore, an increase (decrease) of innovations in a technical field will lead to a growing (shrinking) number of new technical standards in this field. Another reversed feedback process originates from the influence of standards on the innovation activities of companies. On the one hand, too many technical standards paralyse innovative actions, because they do not leave enough leeway for new ideas. On the other hand, technical standards reduce the risks for entrepreneurs in times of rapid technical change and build up a solid framework for new innovations based on the general acceptance of the existing set of technical standards. One traditional way to examine these theoretical hypotheses is to look at selected case studies. A new approach is presented in this paper by running a statistical analysis based on aggregated empirical data.

The Data : In a first step the role of technical standards in the technology indicator system will be identified and evaluated. Secondly, the data base, the PERINORM edited by the German, French and British standards development organisations, is discussed. After evaluating strenghts and weaknesses in relation to the use of patent statistics, different indicator concepts are presented. On the one hand, the standards data allow to build up time series of standards publications differentiated by the international classification of standards (ICS) for the period beween 70 and 96. Second, it is possible to calculate the average duration of standards and survival functions. In a further step, a concordant classification between the international classification of patents with ICS was constructed. The patent data contains the patent applications in Germany from 70 to 96.

The Methods : Two types of statistical analysis will be undertaken. First, in a cross-sectional analysis the quantitative relationship between patent applications and standard publications is examined. Second in a longitudinal analysis, time series of patent applications and standards publications will be investigated, looking for significant relationships and time lags.

The Results : In the cross-sectional analysis, it turns out that there is a positive correlation between patent applications and standard publications, especially if environement, safety and health standards are ignored. This supports the results found for the UK by Swann et al. (Economic Journal (96, p. 1307). The expected negative, but not significant correlation is found between the number of patent applications and the average life time of standards. The cross correlation procedure in the context of a time series analysis shows in many cases that the numbers of patent applications are a significant positive leading indicator for the number of standards published with an average time lead between zero and five years, while the effect of standards on patent applications could not be detected. This means that technical innovations will lead to corresponding publications of standard documents by the institutional standardisation process. However, a positive or negative influence of the set of standards on innovation cannot be proved.

Conclusion : This first approach to examine the relationship between technical innovations measured by patent applications and technical standards published by authorised standard development organisations in general supports quantitatively the hypothesis of a relationship. Looking at time series, the SDOs apparently pick up the technical change by publishing the corresponding new technical standards. However, the feedback loop of the systems of technical standards on innovations and technical change has not been statistically proved yet. Further research efforts have to be undertaken in this area.

KNOWLEDGE-BASES AND INTERACTIONS IN THE NORWEGIAN SYSTEM: A PATENT-SHARE AND CITATION ANALYSIS. STEP-REPORT

E.-J. Iversen
The Stepgroup - Storgata 1
0155 OSLO  (Norvège)

This article is part of a comprehensive project to 'map Norwegian knowledge bases' for the Norwegian Research Council; as such the use of patent-data complements other quantitative (CIS, bibliometric, input-output) as well as qualitative (industry-based case studies) approaches. In this context, Norwegian patenting in the US was used to proxy technical knowledge-creation while citations from these Norwegian patents were used to indicate interaction between knowledge-bases. The first section consists therefore of a straight-forward patent-share analysis in which Norweigan patenting (1990-1996) is indexed to the total population to suggest patterns of specialization. Against this background, the knowledge-interaction section looks at two types of interaction. First it explores technological spillovers as traced by first page patent-citations (by primary class). Secondly, it investigates technology-science links, as testified to by patent-citations to Non-patent-literature, mainly journals. In this way we measure dimensions of the main Norwegian knowledge bases as revealed by the patent-lens, as well as the stregnth of the main interactive links between such bases. In doing so it supplements the other sources investigated by this 'mapping project'.

INTERNATIONAL PATENTING AND PATENT PROTECTION

W.-G. Park
American University
Dpt of Economics
4400 Massachusetts Ave
NW D.C. 20016 Washington (USA)

J.-C. Ginarte
Central Europe Department
World Bank
Washington D.C. (USA)

International patenting data indicate where new inventions come from and where they go. While previous work has emphasized the importance of inventions to national productivity and international competitiveness, it has produced little evidence to date on the effects of patent protection on international patenting activity. This paper uses an index of patent rights in a panel data set of 16 source countries (where patent applications come from) and 40 destination countries (where patent applications go), from 1975-90, to determine the importance of patent protection and other relevant variables on the international diffusion of new ideas. The sample consists of both developed and developing economies. The results indicate that the level of patent rights and the size of the market in destination countries are positively significant influences attracting new foreign technologies. However, much of the variation in the international patenting data (about 70-80 %) are attributable to fixed country effects.

NEW INDICATORS FROM PATENT DATA

D. Guellec, B. van Pottelsberghe de la Poterie
DULBEA
CP-140
50 av F.D. Roosevelt
B-1050 Bruxelles (Belgique)

The purpose of this paper to design alternative weighted counts of patent granted by the European Patent Office (EPO). As a first step, we distinguish between the number of patents filed and patents granted, with respect to both their priority date (date of first application in a national patent office) and their application date at the EPO. These series are further analysed according to qualitative criteria (such as the number of associated technical classes, the designated states for protection, and the number of countries implied in the research process). The second step aims at nowcasting the number of patents granted on the basis of patents filed in recent years. In this respect, we want to test the extent to which the probability that a patent is granted is sensitive to its technological class, the origin country, the designated states, and the technical complexity of the patent. In a similar vein, the lag between the time a patent is granted and the time it is filed is to be analysed with respect to the same indicators. The third step consists in evaluating, both at the macro and industry levels, the effectiveness of these indicators in explaining total factor productivity growth, labor productivity growth, and R&D investment growth.

The paper is structured as follows. Section 2 summarizes the main advantages and drawbacks of patent data as indicator of technological change. Section 3 presents alternative series of patent counts differentiated according to various criteria. Section 4 concerns the econometric analysis of the "grant" probability and of the lag time between the application and the grant of the patent. In section 5 the validity of these new indicators, and especially the nowcasted count of patent granted, is tested with respect to indicators of economic efficiency. Section 6 concludes.

INNOVATIONS AND PATENTS IN BULGARIA

R. Chobanova
Institute of Economics
Bulgarian Academy of Sciences
3 Aksakov St.
BG - 1040 SOFIA (Bulgarie)

A small part of innovation, estimated for the various countries between 10 and 20 per cent, and corresponding to strictly defined and to a great extent specified requirements, receive a patent legislative protection, i. e. the exceptional rights over them and for their usage belong to the patent owner. Responding to the legal requirements for world novelty, innovation level and industrial applicability, the patent inventions or useful models, in their function of an industrial property, can reasonable be regarded as top innovation achievements. The world patent system, which this country is part of, allows an accurate survey and statistics of inventory activities by countries and areas of technics, which could serve as a basis for assessing the tendencies in the general innovation activity in each country, as well.

The protection of industrial property is an important instrument for the industrial development and for the conquering and strengthening of market position under the conditions of a free and competitive economy. Patent activity indicates the interest of companies and organisations in a certain market (or territory), either for the sale of goods, or for investing capitals, or just for the sake of blocking off competition. In this sense, patent statistics could provide useful information about the capacity of the national intellectual potential which is industrially oriented, as well as about the disposition and willingness of foreign capitals to trade and invest on Bulgarian territory. During the past years, Bulgarian and foreign patent-applicants seek protection mostly for their inventions in the areas of: pharmaceuticals and medicines, pesticides, herbicides, etc., heterocyclical compositions, micro-organisms and ferments; medicine, veterinary science; assembly parts for machinery; basic electrical elements; equipment for electrical cable power provision, etc.

Figures show the interest activity in the area of inventions for the past several years. Viewing inventory activities partly as a reflection of the scientific, production and investment policy of a certain country, it is not difficult to account for the slump in effect (measured in times) which could explain the tendencies in the overall innovative activities. These tendencies are typical for economies in transition and they have been evident in countries like Poland, Hungary, the Czech Republic, etc. The exit is to overcome the collapse experienced by science and production, and the above-mentioned countries are going out of that slump at a considerably higher pace than Bulgaria.

The decreased innovative activity throughout the past years is due both to the general economic situation and to the cancelling of the activity of certain mechanisms which, to some extent, favoured the development of innovative processes in this country.

DU VEILLEUR A L'EXPERT DE L'ENTREPRISE AU LABORATOIRE : POUR UNE VEILLE TECHNOLOGIQUE ACTIVE

P. Faucompre, L. Quoniam, H. Rostaing
CRRM - Université Aix-Marseille 3
Case 161 Centre Saint-Jérôme
13397 Marseille Cedex 20

La veille technologique ne peut pas s’appuyer sur une méthodologie de recherche documentaire classique car celle-ci pose seulement des questions précises et en attend des réponses précises. L’approche innovante de la veille réside précisément dans cette rupture de perspective. Aussi, pour le veilleur, l’émergence d’informations nouvelles et le suivi d’indicateurs d’alerte présupposent-ils de traiter d’importants volumes de données éparses et hétérogènes. Cependant, la diffusion et la généralisation récentes des outils et des techniques de surveillance sont telles que de nombreux acteurs industriels disposent aujourd’hui de capacités d’observation et d’analyse très similaires. Ces traitements ne peuvent donc plus prétendre assurer à eux seuls un avantage compétitif qui serait exclusif. En conséquence, les bénéfices issus de la veille technologique, qu’ils soient sous la forme d’avancées technologiques sectorielles ou sous la forme des externalités négatives directes, celles que l’entreprise pouvait espérer infliger à ses concurrents immédiats, risquent ainsi à terme de s’amenuiser peu à peu.

Pour échapper à cette usure accélérée, la veille ne peut donc plus se limiter à la simple surveillance de grandes masses d’informations ou de leur flux. Elle doit adopter une méthodologie beaucoup plus active dans laquelle il ne s’agirait plus seulement d’observer ces informations mais également de provoquer de nouvelles relations entre elles. Partant de cette analyse, nous avons conçu un système de liens automatiques entre banques de données scientifiques et banques de données techniques. Ce système a été appliqué à une base régionale de données bibliographiques de type scientifique afin de

les mettre en relation avec des références bibliographiques de type brevet. Le nouveau profil de cette base, explicitement orientée vers les données de brevet, la destine donc principalement aux entreprises impliquées dans des actions d’innovation technologique et aux entreprises engagées dans la recherche de solutions techniques ou de compétences plus fondamentales telles qu’on les trouve habituellement dans les laboratoires de recherche scientifique.

L’exploitation de cette base apporte deux principaux enseignements. En premier lieu, l’observation des résultats obtenus met en évidence l’intérêt qu’il y a de renoncer à vouloir établir, entre deux univers si différents, l’information scientifique d’un côté et l’information technique de l’autre, des correspondances exactes de type strictement documentaire. Leur analyse démontre comment l’utilisation des liens très larges permet l’émergence de relations qui ne sont pas prévues à priori et qui, par définition, ne peuvent pas être proposées par des systèmes isolés. Ces relations doivent être prudemment interpréter comme de simples présomptions de lien et devenir ou sujet d’étonnement de la part du veilleur ou objet d’analyse pour l’expert. En second lieu, une approche volontairement dynamique et productrice de nouvelles informations permet d’enrichir une méthode de simple surveillance. Notre démarche consiste donc à développer plus largement l’activité de la veille technologique, considérée comme chaîne de traitements, en lui ajoutant une phase préalable de production de liens. Cette première phase active ne se substitue pas aux deux phases d’observation et d’analyse mais a pour effet de les déplacer en aval. Quelques exemples de relations science-technologie illustrent cette double démarche centrée sur l’information brevet et montrent quels avantages potentiels pourraient en être attendus.

UN ESSAI DE CLASSIFICATION TECHNOLOGIQUE DES BREVETS DEPOSES EN TUNISIE :
UNE APPROXIMATION DE L'ACTIVITE INVENTIVE DURANT LA PERIODE 1894-1994

S. Ghali
Faculté des Sciences Economiques et de Gestion de Tunis
Campus Universitaire
Boulevard du 7 novembre
1060 TUNIS (Tunisie)

A partir des caractéristiques principales du brevet, à savoir l’exclusion lorsqu’il s’agit d’accorder à un seul individu ou une seule entité la propriété d’une invention, et la diffusion du savoir lorsqu’il s’agit d’informer sur l’état des connaissances dans un domaine précis, notre objectif, dans cette étude, sera d’étudier les brevets déposés en Tunisie entre 1984 et 1994.

Dans un premier temps il s’agira de faire ressortir les qualités et les lacunes du brevet en tant qu’indicateur de l’activité inventive et innovatrice, puis dans un deuxième temps de procéder à la classification technologique des brevets déposés en Tunisie, qui est absente de la banque de données de l’I.N.N.O.R.P.I (Institut National de la Normalisation et de la Propriété Intellectuelle), tout en étudiant le profil des déposants ainsi que la ‘’ qualité ’’ des brevets

L’objectif de cette étude, en considérant les brevets comme un stock de connaissances à la disposition de l’économie, sera de procéder à une classification par secteurs et produits des brevets qui ont été déposés en Tunisie durant la période 1984-1994. La deuxième étape sera d’exploiter les résultats ainsi obtenus pour profiter de l’une des fonctions du brevet, à savoir un indicateur de l’activité inventive et innovative, pour procéder à l’analyse de la situation prévalent dans ce domaine au sein de l’économie tunisienne. Il en ressort quelques conclusions assez intéressantes dont notamment :

- La majorité des brevets déposés sont d’origine étrangère.

- Ces brevets sont concentrés sur quelques secteurs, et quelques produits.

- Il y a une certaine concordance entre la part des brevets originaires de certains pays, et la part de ces pays dans les flux commerciaux vers la Tunisie.

Quant à l’analyse des brevets déposés par les nationaux, ils obéissent à la tendance internationale quant à leur concentration sectorielle, mais l’analyse du profil des déposants révèle un problème important, c’est le fait que la majorité d’entre eux sont déposés par des individus. Cette situation est le reflet de problèmes structurels vécus par les entreprises tunisiennes, ce qui pose le problème de leur capacité à faire face à la concurrence étrangère dans le cadre des accords du G.A.T.T et de la zone de libre-échange avec l’U.E, d’autant plus que les dépenses en R&D ne représentent que 0,29% du P.I.B (1993) dont à peine 6,8% sont réalisées par les entreprises.