The immense wealth being accumulated by U.S. technology companies and their owners has been apparent for some time, and events during and since the last presidential election have put this reality firmly in the spotlight. Wealth is power, and innovative data practices have allowed for a great concentration of that power among a few key companies and individuals. In Valuing Social Data, Amanda Parsons and Salomé Viljoen provide a timely analysis of this new market reality and help us to think about how our legal systems might better respond. Their article is timely and incredibly useful both for those new to thinking about the data economy and for those looking for new frameworks to address wealth and power disparities in modern society.
Parsons and Viljoen’s article is situated within a broader literature addressing the challenges created by the collection, use, and sale of data in today’s world. Companies operating in this new economy have been able to obtain powerful market positions both through their innovation and by operating outside the scope of existing regulatory regimes—tax systems included. Parsons and Viljoen explain that issue and provide useful terms and taxonomies to better understand and discuss potential responses.
One thing that the authors do particularly well is to demonstrate how tropes like “data is the new oil” fail to reflect the unique attributes and potential harms of data. Data is a valuable commodity largely because those who collect it can combine and analyze it in connection with other data, allowing informed inferences about groups of people. The authors use the term “social data” to refer to this aspect of the data economy.
Defining “social data” is necessary to their analysis, but it is largely instrumental toward a better understanding of how that data is valuable. When we think of value in the tax context, we often discuss exchange value, or the value of an asset in a market exchange. But value can, of course, mean other things. For example, value can also be discussed in ethical or sociological terms. What do we value? What values do we hold?
Parsons and Viljoen introduce these distinctions before specifically focusing on what they call “prediction value,” which is the value derived from the ability to predict future behavior. For example, by tying together purchase history, location information, and medical history, a company may be able to predict a consumer’s preferences or needs in ways that offer the company a competitive advantage. And companies may also be able to predict shifts in group behavior or desires as well, providing them with even greater market opportunities and other sources of power.
Having introduced the concept of prediction value, Parsons and Viljoen then outline three common “scripts” that companies use to extract the prediction value of social data. The first two scripts involve actions that are the type most discussed in the tax literature—selling targeted advertising and generating new revenue streams through product development and innovation. Those scripts transform prediction value into exchange value and provide monetary wealth to the companies using them.
The third script is slightly different, though, and provides the foundation for what, to me, were the most thought-provoking aspects of their article. That final script involves companies using data to cultivate and retain market power rather than to generate current cash flow. For instance, the third script could include the adoption of “strategies of innovation focused on rentiership” or the “use [of] the power cultivated via prediction value to evade or influence regulation.” (Pp. 1037-38.) Parsons and Viljoen argue persuasively that this unique script merits consideration separate from discussions about the power represented by exchange value and monetary wealth.
The foundation of Parson and Viljoen’s article is in providing these useful lenses through which to think more carefully about the data economy, and their article does so very well. The article’s observations regarding the third “script” also invite creative thinking about whether and how our legal systems can or should respond to the distinct issues created by the accumulation and non-market-exchange uses of power by those who possess social data. The final part of Valuing Social Data starts that process.
Specifically, the authors explore how prediction value “collides” with existing legal systems. Looking at the tax system and related literature, Parsons and Viljoen discuss how a focus on exchange value has led to gaps in how scholars think about taxing the digital economy. In particular, scholars and policy makers in the U.S. and abroad have struggled with how to apply existing taxes in situations where data are not directly monetized. (Consider, for example, the barter exchange of one’s data for access to online services.) Global conversations about the proper allocation of taxing power in the digital economy have similarly been impacted by the space that exists between prediction value and exchange value. Parsons and Viljoen invite new approaches informed by this disconnect, including more fundamental adaptations like taxing data collection itself rather than trying to modify existing tax instruments to fit this new market.
A final aspect of Valuing Social Data that makes it particularly compelling to readers is that their analysis is not specific to tax. The authors use tax as an example of where a better appreciation of social data and of prediction value might help to facilitate needed reform, but they also discuss data and privacy regulation, demonstrating how tax issues both mirror and interact with issues faced by multiple legacy legal systems. The failure of existing legal structures to adapt to prediction value has had widespread consequences, and Parsons and Viljoen provide us with useful tools for thinking about how to respond and for thinking about the interaction between tax and broader social movements.
