ABSTRACT
This paper analyses the underlying reasons why innovators do not apply for trademarks for all of their valuable inventions. Using a unique database of UK innovations linked to innovative firms, the empirical analysis highlights the many ways that firms can alleviate information asymmetries and the constraints imposed by collaborative innovation without taking recourse to trademarks. When information asymmetries are not at stake, i.e. when firms use an already existing trademark for their innovations or when they use intermediaries for its distribution, trademarks no longer serve their purpose, leading firms to avoid using it for their innovations. Open innovation also decreases the incentive to trademark, especially when the innovative process involves users, mainly because of property rights issues or because the innovator prefers to use the clients’ own distribution channels.
Acknowledgments
We are grateful to the UK IPO for allowing us use of their proprietary SIPU data and to David Humphries and Pippa Hall for comments on earlier draft versions of the paper. The views expressed in this paper are the authors’ views and we remain responsible for any errors that remain.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Firms that reported a valuable innovation were asked ‘Did you apply for a trademark for that innovation?’, followed by ‘Did you apply for a patent for that innovation?’. We found only 35 firms of 277 had applied for both.
2 The results are robust to the use of less fine-grained industry dummies (we also ran our models with only 7 macro-industry dummies), in order to avoid the risk of over-fitting of the model.
3 For 18 firms who innovate and did not apply for a trademark we could not perform our probit analyses. This is because some firms indicated different reasons for the decision not to apply for trademark (not included in our three categories), while a few firms did not answer to all of the questions of the survey, resulting in some missing values among our covariates.