Innovation in Creative Industries: Does (Related) Variety Matter for the Creativity of Urban Music Scenes?

abstract

This article investigates the relation between different forms of (related) variety found in urban music scenes and innovation in music. While related variety has been found to be positively associated with several indicators of regional economic development and technological innovation, it remains unclear whether its merits also benefit innovation in creative industries. As innovation in creative industries is based on symbolic knowledge, the degree of variety in local contexts may affect the creativity of artists differently than the innovativeness of engineers and scientists. To test whether specialization, unrelated variety, or (semi)related variety is linked to innovation in creative industries, this contribution applies the concept of related variety to the context of urban music scenes. As innovation in creative industries is hidden from traditional innovation data, we utilize volunteered, geographic, and user-generated information from the social music platform last.fm. From relatedness measures between music genres, we generate our own classification system of music, which is used to calculate different related variety metrics of music scenes. Furthermore, our database allows for the identification of innovation in music as the emergence and combination of music genres. The results of this article suggest that semirelated variety promotes innovation in music, while related variety is only positively associated with combinatorial knowledge dynamics. Additionally, specialization limits innovation in music scenes. Hence, policy concerned with creative industries needs to analyze not only aggregate data but also the composition of regional symbolic knowledge bases.

Key words:

• symbolic knowledge
• related variety
• innovation
• music
• creative industry

Acknowledgments

The authors would like to thank Bjørn Asheim, Ron Boschma, and the participants of the Young Economic Geographers Network (YEGN) workshops for comments and suggestions. Furthermore, we are indebted to all music fans that provided tags to the music platform last.fm. The detailed and constructive comments of two anonymous reviewers, and from editor Jane Pollard, are gratefully acknowledged.

Notes

¹ Since a site relaunch in 2015 that occurred after data collection, six tags are displayed.

² The process of sampling artists and tags is as follows. First, the top one hundred artists for all relevant city-related tags were collected. These artists are also tagged with other, genre-related tags, which were included into the database. For each tag, the frequency of artists in the database with that tag was recorded. Starting with the most frequent genre-related tag, we identified those artists who belonged to the top one hundred artists for these tags and originated from the sample cities. Of course, these artists are also tagged with other genre-related tags, which were then included in the database as well. The identification of artists for genre-related tags was repeated for all tags used to identify more than twenty artists in the database.