Blog

The following is the script that I wrote for my presentation at AlgoSoc 2025: The Future of Public Values in the Algorithmic Society in April 2025. The actual title of this talk was, “From efficiency to economic justice: contesting state articulations of the value(s) of computational infrastructures” – but in talking to folks after the presentation this phrasing of “conceptual impoverishment” seems to have resonated more strongly. Perhaps in a future paper I will further develop the connection to “economic justice” a bit more. 

~

In his book, “The Whale and the Reactor: A Search for Limits in an Age of High Technology” from 1988, Langdon Winner highlights the importance of the politics of language as it relates to our technologies: 

“In debates about technology, society, and the environment, an extremely narrow range of concepts typically defines the realm of acceptable discussion. For most purposes, issues of efficiency and risk (or some variant of those) are the only ones to receive a thorough hearing. Any broader, deeper, or more perplexing questions are quickly pushed into the shadows and left to wither. How is it that we have gotten stuck packaging some of the important issues that face humanity in such conceptually impoverished terms?”

In this talk, I will reflect a bit on the possibility of contesting these “Conceptually impoverished terms” through which we evaluate sociotechnical systems…

To do this, I will share some findings from my research on a swathe of U.S. federal regulations, from roughly 2010 and still unfolding today, designed to enable standardized data sharing across the U.S. healthcare industry – or “data interoperability”. Specifically, I will explore how the state articulates the public value of these data sharing systems, and reflect on how we might contest these particular frames or claims of public value. 

This is the part that I would like to discuss more with this group: how do we not only critique claims about the social or economic value of particular technologies, not only reveal their conceptual impoverishment – but create grounds for collective, democratic contestation?

This work is in close dialogue with other scholars, some of whom are at this conference, thinking about articulations of value and regimes of justification around sociotechnicalsystems, including in the healthcare space specifically…

• Gross, N., & Geiger, S. (2023). Choreographing for public value in digital health?. Big Data & Society, 10(2), 20539517231220622. 
• Sharon, T. (2018). When digital health meets digital capitalism, how many common goods are at stake?. Big Data & Society, 5(2), 2053951718819032.
• Van Dijck, J., & Poell, T. (2016). Understanding the promises and premises of online health platforms. Big Data & Society, 3(1), 2053951716654173.

In a moment I will outline three major articulations of the public value of these data sharing policies and infrastructures in the US. These policies have spanned Obama, Trump, and Biden presidencies, and so some of these articulations of value come from the executive office…for the most part, however, these claims are made by the department of Health and Human Services, which oversees our public insurance programs and the governance of health data standards, requirements, etc.

I won’t get into the details about these changes, but suffice to say that data in the U.S. is highly fragmented across the market, and these policies are effectively reconfiguring the data that is available to doctors, hospital systems, to insurance companies, and to the state itself…It does this partially through the creation of a nation-wide health data governance framework, enabling sharing across organizational boundaries, and partially via the required implementation of standardized health APIs. (Broadly, you can think of this shift as something similar to the governance and technical infrastructures that enable financial information to be exchanged across countries, or from one bank to another).

 

The U.S. has spent over a decade pushing for and investing in interoperable data infrastructures. How is the VALUE of these new infrastructures articulated by policymakers, politicians? 

The most common way of framing the value of these efforts is in terms of creating an improved user experience for doctors and patients. The US healthcare system is highly fragmented, across private insurers and hospitals, and if you see a specialist or switch insurers or go to an ER in a different state, your doctors might not have access to the records they need. I think this is the most common articulation of value because it connects clearly and directly to everyday lives experiences of healthcare in the U.S…

Now, if you dig in a little bit, I think there is some room for contestation here inasmuch as doctors already tend to find themselves OVERWHELMED with the amount of data in a patient record, and don’t have the time to comprehensively review pages of documents…the solution to this problem actually requires careful consideration of practices of care, workflows…well, it requires considering USER EXPERIENCE, not just *more data* as an end in and of itself. But nonetheless, I think this is the justification of these infrastructures which points to something other than efficiency or risk, which points to something like improving CARE….. 

The other major justification for interoperable data infrastructures is to enable a new, competitive market, an “ecosystem of apps”…According to policymakers, by increasing “data liquidity”, the market is opened up to new competitors, who can create new products and services… This is assumed to benefit the patient and provider because…competitive markets mean, again, an improved user experience, improved VALUE of the products and services. 

Finally, although this is not often the headline on public-facing federal websites, among policymakers, politicians and industry folk, another key justification or motivation for interoperable data infrastructures is to enable a fundamental shift in the way that care is paid for in the US… This shift is called “value-based care”, which describes attempts by the federal government to redesign healthcare markets away from paying for CARE – so, paying for doctors visits and surgeries – towards paying for HEALTH, for specific healthcare OUTCOMES…a shift from paying for inputs to paying for outputs, if you will…

This market reform has been a multi-decade effort; some of these payment reforms were included in the 2010 Affordable Care Act under Obama. These new payment models are imagined to help reduce healthcare costs in the US, which are notoriously HIGH relative to other OECD countries, by making healthcare providers more financially accountable for healthcare. (The mental model here is: costs are high because providers are just giving you too much care, willy nilly…if they only get paid when your health improves, then the financial incentives are aligned to an efficient production of health outcomes.) 

Data has an important role to play in this market redesign…it is easy to quantify and measure INPUTS…And all that much informationally harder to capture and quantify health OUTPUTS…did your diabetes improve after a year? If it didn’t, is it the doctor’s fault or maybe it’s because you got in a car crash and that needs to be accounted for?….Standardized data sharing across organizations is advertised as being essential to this transition to quantifying and paying for health outcomes, enabling a more rational and efficient market. 

So, this is how the state articulates the value of these data infrastructures: for the most part, this value is articulated in economic terms, in terms of efficiency and competition and the implied benefits for “users” or “customers”… 

To make a normative statement, I think this reveals a conceptually impoverished way of accounting for the value of data sharing infrastructures. And not only that, but this prioritization of efficiency and competition are reflected in the way that these policies are made and these infrastructures are built…other kinds of public value are not prioritized in these design processes. 

Likewise, when I talk to people about my research, about these data sharing policies, most people’s gut reaction is to be concerned about the potential risks to privacy and consent. These aren’t *unimportant* but it limits the scope of contestation to relatively technical enactments of privacy/consent. The broader question about what these data sharing infrastructures are FOR, whom do they BENEFIT, is largely left unexamined. 

Especially when it comes to new digital technologies, there is often an assumption that these systems will create some efficiency, some productivity…or perhaps that they are neutral and necessary “updates”. To be clear, I am not saying that standardized health data sharing is bad: I think it has potentially immense public importance. But I think it is troubling that the terms of evaluation are in these terms, highlighted by Winner, of efficiency and risk. 

So, what is to be done? How can we intervene in these conceptually impoverished frames for evaluating health data infrastructures, or other sociotechnical systems? How can we create space of public contestation, to make space for articulations of what we want from a data sharing infrastructure, in terms other than “efficiency” and “reduced risk”? 

This is not a rhetorical question. In the U.S. in particular I think these ideas have immense caché, they are hard to question. The popularity of ideas of “efficiency” and productivity has helped to give rise to our new “Department of Government Efficiency”, designed to tackle “waste fraud and abuse.” …

Of course it also doesn’t help that the infrastructures I am talking about here are INCREDIBLY BORING: healthcare payment models, data infrastructures…these infrastructures aren’t sexy, and their importance is opaque…and this is why it is quite easy for figures like Elon Musk and others before that to tinker with and reconfigure these infrastructures with relatively little public contestation. Nonetheless, I think that spelling out some of these claims of value, drawing out the implications of optimizing purely for “efficiency” for healthcare workers and patients, might help to create an opening for a broader set of values – such as “economic justice”, for instance. 

If one of the major claims to the value of these infrastructures relies on the idea that healthcare providers are WASTEFUL and this newly liquid, standardized data will enable increased “accountability” of care, perhaps by having conversations with healthcare providers about how these new flows of data are leading to greater intensification of their work, we can draw out the human implications behind a term like “efficiency”.

To conclude, I wonder if we can discuss together: What does it look like for scholars to build alliances, to enable more public forms of contestation about the public value of new technological systems? I am interested in speaking more with folks about the sites that you study, and what opportunities you see for drawing out the human implications of “conceptually impoverished” frames of evaluation, and of building alliances to contest those claims.