Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Innovation with and without patents - an information theoretic approach

Taalbi, Josef LU (2025) In Scientometrics
Abstract
Patents are among the most common proxies for innovation, and a long-standing discussion is therefore to what extent patents can be used to monitor trends in innovation activity. This study proposes a new straightforward approach to measure the amount and quality of information content of a data source, where information is approached as a transmission from a source to a receiver through a noisy channel. This framework is applied to measure information about actual innovations contained in the patent system, and gives empirical estimates based on a database matching 4,460 Swedish innovations (1970-2015) to international patents. The results show that most innovations were not filed for patent and that among those that were, 43.9% of all... (More)
Patents are among the most common proxies for innovation, and a long-standing discussion is therefore to what extent patents can be used to monitor trends in innovation activity. This study proposes a new straightforward approach to measure the amount and quality of information content of a data source, where information is approached as a transmission from a source to a receiver through a noisy channel. This framework is applied to measure information about actual innovations contained in the patent system, and gives empirical estimates based on a database matching 4,460 Swedish innovations (1970-2015) to international patents. The results show that most innovations were not filed for patent and that among those that were, 43.9% of all innovations, a small fraction can be identified with patent quality data. The estimates suggest that overall patents capture 15% of all information about innovations, equivalent to an information loss of 85%. The overlap between the patent and innovation systems is hence more modest than is often assumed. This accentuates the need to develop versatile and multidimensional approaches, based on both patents and other indicators, in order to monitor and stimulate various aspects of innovation. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
epub
subject
in
Scientometrics
publisher
Akademiai Kiado
ISSN
1588-2861
DOI
10.1007/s11192-025-05406-y
project
SWINNO 3.0 Significant Swedish technological Innovations from 1970 until now
language
English
LU publication?
yes
id
5a6802de-9018-4b5e-8559-1d2764b63450
date added to LUP
2025-08-05 21:05:00
date last changed
2025-08-28 08:40:55
@article{5a6802de-9018-4b5e-8559-1d2764b63450,
  abstract     = {{Patents are among the most common proxies for innovation, and a long-standing discussion is therefore to what extent patents can be used to monitor trends in innovation activity. This study proposes a new straightforward approach to measure the amount and quality of information content of a data source, where information is approached as a transmission from a source to a receiver through a noisy channel. This framework is applied to measure information about actual innovations contained in the patent system, and gives empirical estimates based on a database matching 4,460 Swedish innovations (1970-2015) to international patents. The results show that most innovations were not filed for patent and that among those that were, 43.9% of all innovations, a small fraction can be identified with patent quality data. The estimates suggest that overall patents capture 15% of all information about innovations, equivalent to an information loss of 85%. The overlap between the patent and innovation systems is hence more modest than is often assumed. This accentuates the need to develop versatile and multidimensional approaches, based on both patents and other indicators, in order to monitor and stimulate various aspects of innovation.}},
  author       = {{Taalbi, Josef}},
  issn         = {{1588-2861}},
  language     = {{eng}},
  publisher    = {{Akademiai Kiado}},
  series       = {{Scientometrics}},
  title        = {{Innovation with and without patents - an information theoretic approach}},
  url          = {{http://dx.doi.org/10.1007/s11192-025-05406-y}},
  doi          = {{10.1007/s11192-025-05406-y}},
  year         = {{2025}},
}