Firm strategies and path dependencies: an emerging economic geography of industrial data

ABSTRACT

This paper analyses how firms engaged in industrial data production in the power-generation industry operating within the US energy sector produce, collect, store, aggregate, analyse, and ultimately monetize and use industrial data. Industrial data activities are increasingly central to the strategic decision-making of firms in industries from energy to aerospace. This study focuses on whether industrial data production is dispersed and distributed or geographically concentrated. And further, what factors influence the location and concentration of industrial data-production activities. The results indicate that regional factors are affecting industrial data production and that regional policy may not only influence but also largely determine the evolving economic geography of industrial data.

KEYWORDS:

• economic geography
• regional policy
• industrial data
• competitive advantage
• innovation systems

JEL:

• O
• O1
• R
• R11

ACKNOWLEDGEMENTS

The authors acknowledge the assistance of the research team at the Center for Urban Innovation, specifically Dr Thomas Lodato, Emma French and Chris Thayer, for help with the project. They also thank the research team at Georgia Tech's Strategic Energy Institute's Energy Policy Innovation Center for comments and suggestions.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

ORCID

Jennifer Clark http://orcid.org/0000-0001-7908-032X

Supraja Sudharsan http://orcid.org/0000-0001-5279-4189

Notes

1. We collected data on 60 industrial data enterprises. In some cases, these were different divisions within a single parent firm using different kinds of industrial data for different purposes. Given the heavy participation of large, transnational corporations in production and the use of industrial data, this accounts for the variation between the number of firm cases and number of independent firms studied.

2. In scoping this study to ‘industrial data’, we differentiated it from the development of products and services using consumer data – that is, data from consumer devices or those that derive from the characteristics and behaviours of individuals. The rationale for this distinction rests on several points. First, as data increasingly become a discrete product, their production, use, and value remain varied across sectors and variable across time and space. For example, applications such as smart cities, consumer internet-of-things, health analytics and financial technology have substantial individual privacy and security concerns influencing value, trade and use. Privacy and security concern and accommodates operate differently for industrial data than consumer-oriented data. Considering a ‘person’ (consumer) from whom the data were collected as the unit of analysis complicates the understanding of the production process. By restricting the analysis to an industry where industrial machines are reporting data related to their characteristics and operations, the research can focus on the same technology, but with data that are proprietary rather than private. In other words, the rights to use are contractually and legally defined rather than ethically constrained. In this way, we can simultaneously isolate the strategic decisions made by firms about the production and commodification of industrial data, and analyse the locational factors influencing those decisions.

3. Networks such as the Industrial Internet Consortium and Smart Manufacturing Leadership Coalition enable collaboration between firms across the industrial data supply chain.

 

Additional information

Funding

This work was supported by the Strategic Energy Institute at the Georgia Institute of Technology (Industrial Data Project).