Locating Software, Video Game, and Editing Electronics Firms: Using Microgeographic Data to Study Barcelona

ABSTRACT

This paper analyzes location patterns of software, video game, and editing electronics firms (hereafter SVE) in the Metropolitan Area of Barcelona using microgeographic data. These industries are mainly attracted by agglomeration economies look to benefit from skilled labor and spillover effects, and tend to cluster in large metropolitan areas. However, less is known about their detailed location patterns inside these areas. We contribute to the empirical literature by identifying how SVE firms are concentrated in some core areas of the metropolitan area. Our empirical application includes using the Nearest Neighbor Index (NNI) and M-functions, as well as local spatial autocorrelation indicators.

KEYWORDS:

• Software industry
• video game industry
• microgeographic data
• spatial location patterns
• Barcelona

Acknowledgments

This article has benefited from funding from FEDER/Ministerio de Ciencia, Innovación y Universidades (project ECO2017-88888-P), the “Xarxa de Referència d’R + D+I en Economia i Polítiques Públiques” and the research program SGR (2017 SGR 159). We are grateful for the comments received at the 18^th INFER Annual Conference, XLII International Conference on Regional Science, 1^st Catalan Economic Society Conference, 57^th ERSA Congress and 6^th CERS––Engines of Urban and Regional Development, and the suggestions made by R. Boix and E. Coll-Martinez. We also would like to thank two anonymous referees for their valuable comments, and the editor for all the support provided during the editorial process.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes on Contributors

Carles Mendez-Ortega, is a PhD student in the Department of Economics, Universitat Rovira I Virgili

Josep-Maria Arauzo-Carod, is an associate professor in the Department of Economics, Universitat Rkovira I Virgili

ORCID

Carles Méndez-Ortega http://orcid.org/0000-0003-3626-7500

Josep-Maria Arauzo-Carod http://orcid.org/0000-0002-3801-223X

Notes

1. Creative industries are a group of industries linked to creative and high-tech activities that, among others, include activities such publishing, audio-visual, arts, advertising, R&D, and software.

2. Data from the Institut Català de les Empreses Culturals: “Empreses de videojocs a Catalunya 2015” except the data concerning the number of employees, which is from our database.

3. Universities play an important role in the creation of high tech firms, but it is not the only key factor for the success of these industries; it is also the local environment or the innovative milieu (Mayer, 2007).

4. The Universitat de Barcelona (UB), the Universitat Politècnica de Catalunya (UPC) and Universitat Pompeu Fabra (UPF) offer bachelor’s and master’s degrees focusing on video games.

5. An extensive review of this literature can be found in Arauzo-Carod et al., 2010.

6. The SABI database contains a long list of variables, including year of constitution, balance sheets, income, expenditure accounts, number of employees, industry, sales, assets, and geo-referenced location (X–Y coordinates). The data is collected by SABI from the Mercantile Register, where all limited liability companies and corporations are obliged by law to deposit their balance sheets.

7. Data provided by the Government of Catalonia.

8. Concretely, 2 video game firms do not report any headquarters and 15 have missing locational data.

9. Initially, we compiled 19,229 firms from the province of Barcelona (NUTS 3 level) from the SABI. After some filtering, we discarded 11,491 firms with missing data or which were not located in the MAB, and to this dataset (7,738) we added 74 video game firms in order to obtain a final dataset of 7,812 creative firms.

10. The firm level data includes the location, age of the firm, and the number of employees.

11. We have decided not to use econometrics as this paper is part of a broad research project belonging to a PhD dissertation that approaches location patterns of software and video game firms from different perspectives. While this paper adopts Spatial Autocorrection Analysis, Nearest Neighbor Index (NNI), and M-functions, another paper we have written, ran econometric analysis in order to identify the location determinants of entering firms in neighborhoods of Barcelona.

12. The main aim of this paper is not the analysis of the role of real estate prices but the analysis, in general terms, of location patterns (i.e., rather than location determinants). It is obvious that real estate prices are relevant, but their influence over firms’ locations is out of the scope of this analysis. Although real estate prizes may affect all types of firms (even in a different way), in this paper we are mainly interested in spatial linkages among location patterns of firms from different industries.

13. There are several alternative definitions for neighbors (k-nearest neighbors, contiguous neighbors, or distance-based neighbors, among others) but considering that this is an intra-urban analysis in which neighborhoods are quite close to each other in a defined space (Barcelona) we considered that the most appropriate measure is contiguity.

14. The cut-off point of firms’ ages was obtained by the median, whereas for size we considered that firms with more than three employees were not small firms by this industry’s standards (the median size is five employees).

15. This effect on the level of the MAB is due to the large number of firms located in Barcelona and to the proximity of Sant Cugat del Vallès to Barcelona, which are the second- and first-ranked municipalities in terms of the number of video game firms (the two municipalities have more than 80 percent of the video game firms situated in the MAB), producing this Moran’s I value, which is higher than the level for the city of Barcelona.

16. The calculations were made using a 0.05 significance level and 499 permutations.

17. All calculations were made at a 0.05 significance level, using 150 simulations.

18. The M-function graphs (from A1 to A11) summarized in Table 6 are provided in the Appendix.

19. For further details, see “Las Grandes Áreas Urbanas y sus municipios 2015” at http://www.fomento.gob.es/NR/rdonlyres/416CE7FD-A6B0-431D-881B-D1F07664795E/133984/listado__2015_2.pdf.