![Sample our Global Development journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5929/TOC-Subject-Sample-2021-Global-Development.jpg"})
ABSTRACT
The geography of duplicated inventions: evidence from patent citations. Regional Studies. Innovators often claim inventions that turn out to duplicate, at least in part, existing ones. This paper advances the claim that for recent and upcoming inventions, competitive incentives are high, and localized knowledge flows increase the probability of duplication. Therefore, over a brief period of time the probability of duplication is higher at short geographical distance. Conversely, the duplication of less recent inventions is more likely at long distance as a consequence of a lower awareness of the existence of a technology. This claim is supported by coherent descriptive and multivariate evidence using data on patent citation categories from the European Patent Office (EPO).
摘要
复制创新的地理：来自专利引用的证据。Regional Studies. 革新者经常强调创新至少部分是复制既有之物。本文推进晚近与即将来临的创新有高度的竞争诱因，而地方化的知识流增加了复制的可能性之宣称。因此在短期内，在较短的地理距离中复制的可能性较高。反之，较早期的创新复制则较可能发生在距离较远之处，因其对于技术的存在感知较低。我们运用欧盟专利办公室（EPO）专利引用类别的数据之一致的描述性与多变项证据，支持上述宣称。
RÉSUMÉ
Géographie des inventions dupliquées: témoignage de citations de brevets. Regional Studies. Des innovateurs revendiquent souvent des inventions qui s’avèrent être des duplications, tout au moins en partie, de brevets existants. La présente communication soutient que, pour des inventions récentes et à venir, les incitations compétitives sont élevées, et la diffusion localisée des connaissances accroît les possibilités de duplications. En conséquence, dans l’arc de brèves périodes, la probabilité de duplications est plus élevée en présence de distances géographiques limitées. Inversement, la duplication d’inventions moins récentes est plus probable sur de longues distances, du fait d’une sensibilisation moins prononcée sur l’existence de la technologie. Cette revendication est appuyée par des témoignages descriptifs et multi-variables qui font usage de données sur les catégories de citations fournies par l’Office européen des Brevets.
ZUSAMMENFASSUNG
Die Geografie der duplizierten Erfindungen: Belege von Patentzitaten. Regional Studies. Innovatoren nehmen oft Erfindungen für sich in Anspruch, die sich später zumindest teilweise als Duplikate von bereits vorhandenen Erfindungen erweisen. In diesem Beitrag wird die These aufgestellt, dass es für aktuelle und künftige Erfindungen hohe Wettbewerbsanreize gibt und dass lokalisierte Wissensströme die Wahrscheinlichkeit einer Duplikation erhöhen. Daher fällt während eines kurzen Zeitraums die Wahrscheinlichkeit einer Duplikation bei einer kurzen geografischen Entfernung höher aus. Demgegenüber ist eine Duplikation von weniger aktuellen Erfindungen bei höheren Entfernungen wahrscheinlicher, da die Existenz der jeweiligen Technik weniger bekannt ist. Diese These wird durch kohärente deskriptive und multivariate Belege unterstützt, für die Daten von Patentzitatkategorien des Europäischen Patentamts zum Einsatz kommen.
RESUMEN
La geografía de los inventos duplicados: evidencia de citas de patentes. Regional Studies. Los innovadores con frecuencia reivindican inventos que resultan ser un duplicado, al menos en parte, de otros ya existentes. En este artículo afirmamos que para los inventos recientes y futuros, los incentivos competitivos son altos y los flujos de conocimiento localizados aumentan la probabilidad de duplicación. Por consiguiente, durante un breve periodo de tiempo la probabilidad de duplicación es superior a una distancia geográfica corta. Por el contrario, la duplicación de inventos menos recientes es más probable a largas distancias como consecuencia de un menor nivel de conocimiento de la existencia de una determinada tecnología. Esta afirmación se sustenta por la evidencia descriptiva y multivariante coherente de datos sobre las categorías de citas de patentes de la Oficina Europea de Patentes (OEP).
SCHLÜSSELWÖRTER:
PALABRAS CLAVES:
ACKNOWLEDGMENTS
Special thanks to Monica Coffano, Dominique Foray and Markus Simeth for help and support. The authors are grateful to Stuart Graham, Dominique Guellec, Adam Jaffe, Francesco Lissoni, Jacques Mairesse, Paula Stephan and Arvids Ziedonis for their comments and suggestions. All errors and omissions are the authors’ own.
DISCLOSURE STATEMENT
No potential conflict of interest was reported by the authors.
SUPPLEMENTAL DATA
Supplemental data for this article can be accessed at http://dx.doi.org.10.1080/00343404.2017.1280152
Notes
1. Examples of duplication include: the Galileo thermometer in 1592, also claimed by Otto Van Guericke and Giambattista della Porta in 1606, Cornelius Drebbel in 1608, Sanctorio Sanctorious in 1612, and Paolo Sarpi and Robert Fludd in 1617; the discovery of electric light by Sir Joseph Swann and Thomas Edison; the telegraph; the telephone; electromagnetic clocks; the typewriter; the diesel engine; jet propulsion; and numerous others. Similar examples led Merton to the provocative hypothesis that ‘far from being odd or curious or remarkable, the pattern of independent multiple discoveries in science is in principle the dominant pattern’ (Merton, Citation1961, p. 477). For a broader discussion, see Merton (Citation1961) and Lamb and Easton (Citation1985). Constant (Citation1978) and Elkana (Citation1971) revise some of these examples, claiming that in some instances the level of similarity of the inventions involved has been overstated. See Bikard (Citation2012) for a recent literature review.
2. Note the distinction between this type of knowledge, which can differ between two individuals with the same potential to develop a given technology, and the knowledge base needed actually to develop a technology, referred to in the previous paragraph.
3. To be clear: in the context of the analysis in this paper, patent citations are not meant to be interpreted as an indicator of knowledge flows; they are being used to capture the occurrence of duplicated inventions and to study how they distribute geographically compared with a meaningful counterfactual (see the discussion of the empirical model in the fifth section).
4. Before 1982, citation categories are often not reported, while there may be truncation problems in PATSTAT(2010) after 2007.
5. The results are robust to the inclusion of Y citations in the dependent variable or to their exclusion from the sample. As previously noted, X and E citations share the same definition with respect to the relevance of the citation for the novelty of the cited patent. For this reason, these two categories are considered jointly. In analyses not reported but available from the authors upon request, separate analyses were performed for these two types of citations. The results lead to an equivalent conclusion.
6. Similar results are obtained when considering the simple variables or adopting a second-order polynomial transformation.
7. In the construction of the Same applicant variable, all applicant information available within each citing and cited patent families is used. In consequence, it does not account for potential corporate ties between the citing and cited applicants.
8. For instance, if, hypothetically, less experienced inventors with a higher probability of duplication were remotely located compared with more experienced ones, descriptive evidence would be biased upwards. This concern does not apply instead in the model presented here.
9. Not reported, but available from the authors upon request.
10. The values discussed refer to the probability that a citation is XE in our sample, where we only consider patents with at least one XE and one A citations. These values show that the likelihood of observing an XE citation varies significantly relative to the sample average (48%). However, they cannot be interpreted as ratios of duplicated inventions in absolute terms: 48% cannot be interpreted as the percentage of inventions that are duplicated.
11. For the World Intellectual Property Organization’s (WIPO) IPC and technology concordance table, see http://www.wipo.int/meetings/en/doc_details.jsp?doc_id=117672/.