A different way to govern data is possible
An interesting paper explores some of the models that are emerging in response to the current data governance landscape
1The current model of economic exploitation of data is unbalanced. A limited number of corporations (mainly internet platforms and telecoms) seem to control the access and monetization of most of the data that we all generate. This may not be just a byproduct of other technological or economic dynamics but, according to Sadowsky, Zuboff or Srnicek, the result of a model of (data/surveillance/platform) capitalism that pursues data accumulation as its key economic logic (with many pretenders trying to follow suit, not always successfully, to the dominant data barons).
In response to this paradigm of concentrated corporate control over data, a myriad of new models have born that try to provide more democratic and equitable alternatives. In their article Emerging models of data governance in the age of datafication, Marina Micheli and colleagues (2020) reviewed more than 72 academic articles, 16 book chapters, 63 reports and policy documents, and 22 websites of projects/initiatives to design a categorization of these new emerging models. This is a hot topic: three quarters of the reviewed material had been published from 2017 onwards.
The article defines data governance as “the power relations between all the actors affected by, or having an effect on, the way data is accessed, controlled, shared and used, the various socio-technical arrangements set in place to generate value from data, and how such value is redistributed between actors.” It then lays out the five dimensions used to dissect the models: stakeholders, governance goals, value from data, governance mechanisms and reciprocity.
Four main data governance models emerge from their analysis:
Data Sharing Pools
This basically refers to ad-hoc agreements between data holders (corporations, public bodies, etc.) to exchange data through contractual arrangements. The goal is basically to create new products or services that can result in economic benefit for the parties to the agreement.
Personal Data Sovereignty
In this model, data subjects (individuals) retain some control over their data, increasing their ability to decide with whom and under what conditions they wish to share their data. The goal here is to give data subjects greater self-determination over their data, but at the same time, enable them to share it so that the aggregate data can be exploited for business or social benefit.
These two models do try to generate alternative mechanisms to access and control data by individuals and organizations, but do not seem to fundamentally question the value proposition of data as a commodity or an asset that can be aggregated and then monetized to generate economic profit. Although control over data may be de-concentrated and devolved to individuals, the task of using data and generating benefit from it remains in private (mainly corporate) hands.
The remaining two models, by contrast, try to move the central locus of the value generation away from corporate actors.
Data Cooperatives
These are grassroots-driven governance models in which citizens share data on a voluntary and decentralized fashion to a cooperative, which uses the data to achieve a collective value. Here, data is not so much an asset to be monetized or profited from, but a commons that is put to the service of the cooperative’s goals.
Public Data Trust
In this last model, public organizations play the central role. Data is conceived as a public infrastructure whose access and control remains in public hands and is leveraged to generate public value. In this model, it is the public organization who is entrusted by data subjects to collect, process and use the data to achieve that public value.
The typology is interesting and useful to explore this emerging field. The first two types (Data Sharing Pools and Personal Data Sovereignty), however, seem more like concrete governance instruments or tools than data governance regimes. The other two, with their deeper questioning of what is the value that data can generate and how this can be distributed in society (through a cooperative arrangement or a public body) does have more profound implications for data governance at large. At the same time, the paper does recognize that there is limited research concerning the value that these two models generate and about their sustainability.
In practice, most experiments with new governance arrangements show elements of more than one of the four ideal types distilled by Micheli and colleagues, as they explicitly acknowledge in their paper. For example, the data governance model developed in the City of Barcelona has elements from the four. Mostly described as pursuing (personal and city) data sovereignty, the government played the central role in the establishment of the rules and instruments over the access, control, management and use of data. Some of these instruments were data sharing contracts or mandatory data sharing requirements for companies operating in the city. The City also established mechanisms to give citizens greater control over the data they shared. Under the EU funded DECODE project, the City also promoted and tested a technology for decentralized data sharing, where citizens could share data while retaining control over it, all for the purposes established by the community and through distributed ledger and privacy enhancing technologies.
With the current hype around decentralized infrastructures as the building blocks for a new governance of the internet, the case in Barcelona provides interesting insights into some of the opportunities and limits of such models for urban data governance. The typology developed by Micheli and colleagues, and more specifically, some of the elements that they identify (such as the governance instruments or the focus on what value and for whom is generated) provide very useful analytical tools to continue exploring this rapidly evolving phenomenon.
Image Credit: Digistranscope. The governance of digitally-transformed society. A JRC Science for Policy Report available at https://publications.jrc.ec.europa.eu/repository/handle/JRC123362