First considerations towards industrial data sharing practices

Guest writer Timo Seppälä continues the discussion started in May’s RxEvent on how we could benefit from shared data

Data collection, data analysis, and data sharing enable more productive industrial supply chains. The recipe for these more efficient industrial supply chains will no longer be contingent on how different companies contract to share data in one-to-one business relationships, but how these companies make these contracts available for others across the industry. Advancing the availability of data-sharing contracts has led to novel internal and external operational efficiencies and to new types of strategic research and development opportunities. Hence, data sharing is nothing new. Approximately 49% of the companies already share data with other companies across their operations.

Data cannot be owned

As a rule, data cannot be owned. Data may belong to various industry actors, but they cannot be owned in the legislative sense. The most natural view of who controls data access is the industry actor who owns the asset. The ownership of an asset is the default situation of data access management when no contractual arrangements or like have been made. In this case, the owner of the asset usually has a natural ability to control others from accessing the data by pre­venting access to the asset in operation.

On the other hand, every industry actor, such as the asset owner, asset manufacturers, component, and raw material suppliers and any other industrial third party may have their own interest in accessing the data produced by different assets in operation. However, the situations can sometimes involve excluding other industrial actors for accessing data through the life cycle of the asset. In addition, it could be stated that an asset owner has ownership-like administration of data when it can deny other industrial actors the use of the data even when it does not have actual ownership.

When does data become intellectual property?

Another aspect of data ownership is intellectual property rights. On the one hand, they do not concern the accessing data. In general, intellectual property rights can be administered and agreed upon. On the other hand, there is plenty of compelling legislation pertaining to intellectual property rights. Intellectual property rights are often divided into copyright and industrial property rights. Both establish how the rights are utilized in business, but they are also valid for a limited time. From this perspective, intellectual property rights constitute stronger ownership specifically because they contribute to the factual ability to prevent others from using the data indirectly through a court. Additionally, it should be considered that intellectual property rights emerge at the stage when data are used for specific purposes, for example as part of new research and development. Even at present, industry actors administer the new data produced through research and development projects and their intellectual property rights.

A second view on accessing data is the data protection related particu­larly to personal data, i.e., at the stage when the data could lead to the identification of a per­son when the data is personal data and involves certain statutory obligations. An actor who administers data related to a person, i.e., a register of personal data, must safeguard the data as necessitated by data protection. 

Adding data sharing clauses to your agreements including 3rd parties

The third view on data ownership and management is agreements. Within the free­dom of contract, it can be specified who data belongs to, what kinds of access rights there are to the data, whether they are exclusive, parallel, etc. It is aimed at agreements between parties on accessing data and use restrictions even when no one owns the data and only restrictions on any contractual partner. The restriction of contract comes, however, from the fact that the contract cannot be binding on a third party.

From the perspective of contract law, the reliability (or correctness) of data will be an in­creasingly significant element, one which will fundamentally shape contract policies between parties as data are moved through different interfaces between the various actors. However, even in long data transmission chains, it must be possible to contractually establish causality of liability. Even if such direct and indirect legal means of prevention could be created, they would probably not be enforceable everywhere in the world, i.e., it may not be possible to use the local legal system to prevent the users of data from doing so.

The factual management of IPR, data protection, and agreement policies are the three aspects that data accessing data involve and where the present legislation sets certain marginal conditions. In the end, the contractual policies between the actors will define the relative strengths of data ownership and control be­tween parties.

What’s next on data sharing roadmap?

To conclude data should be considered as important and any other production resource, just like human labor, machinery, capital, and other such constituents. Additionally, when planning data sharing categorization of data should be considered into two broad categories: proprietary data and shared data. To elaborate further, shared data can be divided into three subcategories: 1) data shared with trusted partners—or confidential data; 2) data shared with other stakeholders—or distributed (industry) data; and 3) data shared with anyone—or open data. After the categorization of the data, the data-sharing contracts should be considered.

Timo Seppälä Researcher, ETLA & Professor of Practice, Digital Operations, Aalto University PhD (Industrial Economics), M.Sc. (Entrepreneurship)