Innovation's Econometrics : PATENT

Plateau du Kirchberg Luxemburg
November 28 & 29 , 1996


The Key role of patents in Science and Technology

Henir Dou
Université Aix-Marseille III

Patents are a unique link between Science and technology. They even on a larger point of view link Science and Society. The development of innovation within hyper-competitive firms (small of large companies), necessity prior to any type of move a deep knowledge of the technologies (and even more the protected technologies) and the different results of research which will be integrated in new products and services.

Within this landscape, and because most of the results claimed and described in patents are seldomly published elsewhere, the patent databases are one of the most important thinktank available in modern world. We will expose first of all in this paper how the automatic analyze of patent leads beyond knowledge of the interaction of different area of technology and science when the invention covers more than one area of application. Second of all, we will give an example of the database developped in Marseille at the CRRM on internet (http://errm,univ-mrs:fr), from the interdisciplinary PASCAL, database and where the PASCAL's codes have been completdd with the most closer IPC codes. This means that a researcher can from its own work, see that patents exist, and that its own work (or the work of other people) may be related to some industrial applications. The method used in transcoding the data will be explanned.

Various example will be given in both category of applications. If necessary, the automatic patent analysis overview will be completed by a few analysis examples on various possible correlations.

The conclusion will shows how firms must think to ameliorate their product, but also how they must complete this action by looking permanently to the functions of the product and how they are fulfilled by its competitors. The patent databases, with an access by words and IPC codes are one of the best tool available in the world to achieve this fundamental practice.


Les brevets constituent un lien unique entre la science et la technologie. Ils constituent aussi sur une base plus large un lien entre la science et la société. Le développement de l'innovation dans les sociétés hyper-compétitive (pour les petites et grandes firmes) nécessite avant toute action une bonne connaissance des technologies (et sans doute encore plus des technologies protégées),ainsi que des différents résultats de recherche qui seront ensuite intégrés dans de nouveaux produits ou services.

Dans cette perspective, et aussi parce que les résultats revendiqués dans les brevets ne sont que rarement publiés autre part, les bases de données brevets constituent un des plus importants réservoirs de connaissance dans le monde moderne. Nous exposerons premièrement dans cet exposé comment l'analyse automatique des brevets, au-delà des questions classiques qui fait quoi, où, quand, comment ? conduit à une meilleure connaissance des interactions de différents domaines d'applications ou de réalisation de l'invention décrite dans le brevet.

Dans une seconde partie nous traiterons de la base de données développée au CRRM à Marseille ( Cette base de données réalisée avec la base interdisciplinaire PASCAL présente les codes PASCAL transcodés en code CIB Cela veut dire qu'un chercheur pourra voir d'une part que les brevets existent réellement mais aussi à quelle partie de la technologie son travail peut être relié. La méthode utilisée pour réaliser ce transcodage sera exposée.

Différents exemples seront données dans les deux domaines d'application qui sera complétée par quelques exemples complémentaires portant sur différentes corrélations.

La conclusion montrera comment une société doit penser à améliorer ses produits, mais aussi à compléter cette action en regardant de façon permanente les fonctions de ce produit et comment elles sont remplies par ses concurrents. Les bases de données brevets avec leur accès par des mots et par la classification internationale sont un des meilleurs outils pour réaliser cette pratique fondamentale.


The Geographical Structure of Patent Activities : A Patent Atlas

Dr. Siegfried Greif
Deutsches Patentamt
D - 80297 Muenchen

1. Objective

A study will be presented aiming at findings on the geographical structure of patent activities, the inventor´s activities documented thereby, the R&D activities deduced therefrom and the innovation potentials shown therewith. The study relates to Germany.

2. Basis of data

The subject matter are patent applications of domestic applicants with the German Patent Office (GPO), i.e. 35 000 to 38 000 per annum in recent years. With these figures the patent applications of German origin are virtually completely covered.

Database is the internal databank of the GPO, which is much more efficient in volume and up-to-dateness than the published data.

3. Determination of location

The basic data contain informations on the seat of applicants and inventors. The analysis considers both criteria. Selecting the seat of the applicant may lead to certain inaccuracies by several registered seats and by regionally distributed sites of production and research of a firm. The investigation based on the seats of the inventors makes a better identification of the R&D sites possible.

4. Regional units

The basic assignment of patent applications is performed on the level of communities (town, village). Aggregations on the basis of these communities are developed, on the one hand according to administration units on the other hand according to functional units. The results show a very unequal distribution of patent applications. Obvious are a few strong accumulations of patent activities. According to the seat of applicant about 22 per cent of all applications are concentrated on the big cities of Muenchen, Stuttgart, Frankfurt/Main and Berlin.

5. Aspects of time

The investigation comprises static and dynamic elements. A structural analysis is the centre of the study. For this purpose data of the years from 1992 to 1994 were chosen and an average calculated thereof. The three years are not sufficient to show a development but appropriate to eliminate accidental deviations as far as possible. At short term, the flow of inventions resulting from R&D is varying. Analyses of time series are performed on the level of the Laender of the Federal Republic of Germany and in respect of selected regions. From these analyses result specific regional structures of development, for example with regard to the increase or stagnation of patent activities.

6. Differentiations

Differentiations with reference to various aspects are carried out and the distribution of the respective patent applications is being demonstrated, in particular referring to categories of applicants :

Business enterprises, research institutions,free inventors
Technical units and sectors of economy
Firm sizes.

7. Relations

To extend and to specify the findings of analysing the geographic structure of patent applications, patent data are brought together with other data, in detail referring to the following items :

Population : Density of patents measured in applications per capita of inhabitants Employees : Intensity of patents measured in applications per capita of employees R&D personnel : R&D productivity measured in applications per capita of R&D personnel, regionally specific input-output-patterns.

8. Patent Atlas

The result of the investigation is the Patent Atlas for Germany. The findings will be of interest to science, business enterprises and political institutions. They could be the basis for regionally oriented actions, especially with regard to science and technology policy. In addition to that the regionalized patent data may be used as controlling instruments for the effectiveness of governmental politics.


Size and Innovative Activity : The Greek Case

Zoe GEORGANTA, Asteris Hatziparadisis,

Panagiotis Panousis

Numerous studies have been conducted during the last 30 years in order to test the "Schumpeterian hypothesis about the advantages of size for innovative activity. These studies differ according to the time period covered, sample of firms and industries, measures of innovative activity and firm size, sources of data, and methods. Despite the differences, however, a number of common findings have appeared concerning the relationship within industries between firm size and R & D conducted are closely related to the size of the firm, while R & D effort.

Thus, a large number of studies have shown that within industries, the propensity to perform R & D and the amount of R & D conducted are closely related to the size of the firm, while R & D productivity declines with firm size. In spice of many decades of study, a broad consensus has not yet been achieved, as one can observe on the basis of very recent literature.

For Greece, the relationship between size and R & D has never been examined in the past. Thus, although we do not introduce any novel technique, the Greek case may shed some light on policy issues within the framework of the EU technology strategy.

This paper investigates the relationship between firm size, R & D effort, R & D productivity and number of patents for four cross-sections of Greek firms, 1988, 1990, 1992 and 1994. The data come from the R & D File held at the Indicators Division of the General Secretariat of Research and Technology of the Ministry of Development, as well as from the Industrial Property Organization (OBI). Apart from the conventional econometrics used in this field of research, we also attempt to apply the idea of R & D cost spreading over the firm's output.


Un essai de classification des brevets en Tunisie durant la periode 1984-1994.

M. Sofiane Ghali, Professeur a
L’Universite de Tunis III

A partir des caracteristiques principales du brevet, a savoir l’exclusion lorsqu’il s’agit d’accorder a un seul individu ou une seule entite la propriete d’une invention, et la diffusion du savoir lorsqu’il s’agit d’informer sur l’etat des connaissances dans un domaine precis, notre objectif, dans cette etude sera d’etudier les brevets deposes en Tunisie durant la periode 1984-1994.

Dans un premier temps il s’agira de faire ressortir les qualites et les lacunes du brevet en tant qu’indicateur de l’activite inventive et innovative, puis puis dans un deuxieme temps de proceder a la classification technologique des brevets deposes en Tunisie qui est absente de la banque de donnees de « l’Institut National de Normalisation et de la Propriete Industrielle » (INNORPI) depositaire officiel des brevets. Donc il s’agira essentiellement de proceder a une classification sectorielle pour ensuite etudier la repartition par secteur, branche, nationalite, nature (procede ou produit), statut (firmes ou individus), en tenant compte des transformations temporelles de toutes ces caracteristiques.

Cette etude est d’autant plus importante, qu’elle permet d’apprecier l’activite inventive et innovatrice, en l’absence de donnees relatives aux depenses en R&D au niveau des firmes tunisiennes. C’est en quelque sorte un des indicateurs essentiels qui permet d’apprecier les performances potentielles au niveau de la competitivite des firmes, au moment ou l’economie tunisienne s’ouvre de plus en plus a la concurrence exterieure apres avoir signe les accords du GATT et etablit une zone de libre-echange avec l’U.E.


Innovation Projects

QUICKSCAN - sucessful collaboration with European Patent Office

The 100 selected proposals of the INNOVATION Projects action line (see ITT n° 2/96, p. 16, technology validation and technology transfer projects) have all entered their Definition Phase during which the Commission services offer various accompanying measures like project planning workshop, guidebooks and monitoring tools. One such strategic support initiative is a systematic search of all proposals for their novelty as compared with the current state of the art.

First results are very positive, and the new service could prove to be added value for Commission contractors while at the same time raising awareness for the patent system. The economic utility of QUICK SCAN is paired with its use as a qualitative criterion for the selection of truely innovative projects for the second round of funding. This service is primarily geared towards SMEs. According to a study launched by the European Patent Office, two thirds of European SMEs which generate inventions do not have access to the patent system. A truelyt worrying percentage. In addition, the initial and continued education of engineers, managers and the general public alike in patent matters is clearly deficient in Europe vis-à-vis its main competitors. This makes efforts to make better use of the patent system a major concern for European industry and the INNOVATION programmes saw the necessity to test new ways of raising awareness while, at the same time, offering added value to its contractors.

The added value nature of QUICK SCAN is manifold, utilizing the peculiarities of the patent system and its information base. Primarily, patents are a valuable source for technical (and often otherwise unavailable) information. According to expert estimates, 75 % of all technical knowledge is expressed exclusively in patent literature, i.e. cannot be found anywhere else. The extremely fine classification system (120.000 subdivisions) allows for very precise searches, and a qualified analysis of search results can also take into account the market niches, technological trends and the competitive situation. Patents are early indicators of future technologies, and their analysis can avoid investing in research already performed. Most importantly, a search in patent data bases will yield information on which "claims are staked" by others and thus determine niches and territories for making, using or selling new products or processes.

Proposals to the "Innovation Projects" action line come in a multitude of languages and from a broad scientific, technical, and non-technical base, covering virtually all disciplines. The Commission contracted the EPO as a collaborating partner for QUICK SCAN. Its search branch in the Hague with its 900+ multilingual examiners has access to databases with more than 30 million documents worldwide. A contract was set up realizing the "quick" nature of QUICK SCAN in order to use its results at an early project stage, before significant funding is committed.

The latter is the first objective of QUICK SCAN. If it were found that a proposal would be duplicating work of aiming at fields protected by others this proposal could either be withdrawn or the innovation redirected. The second objective is to raise awareness for the IPR protection system in Europe. Often QUICK SCAN results are new and in certains cases even surprising to proposers, and frequently they are overwhelmed by the amount of information provided on technologies, markets and the competition. The third objective is to offer additional arguments to proposers. A QUICK SCAN is independent documentation of the project's novelty over the current state of the art and thus a good sales argument or positive input for loan negotiations, deliberations on possible joint ventures or strategic alliciances.

During the months April and May, all 100 proposals for INNOVATION projects have been assessed. Without having geared the proposal forms specifically to QUICK SCAN. It turned out that the surprisingly high figure of 90 % of proposals could be searched by the EPO examiners. A full study with the results of the first round of QUICK SCANS will be ready before proposals are sought for the next Call in September, so that lessons can be learned from it.