Sarah Lamdan is a Professor at CUNY School of Law and researchers information law and policy. In November, she will publish the book Data Cartels: The Companies That Control and Monopolize Our Information (Stanford University Press).
Lamdan argues that “data analytics” companies are mining, commodifying, and selling our data and informational resources, perpuate social inequalities and threaten the democratic sharing of knowledge. The book calls for treating information like a public good and creating digital infrastructure that supports our democratic ideals.
In this interview, Sarah Lamdan talked to Rafael Grohmann on Data Cartels book.
DIGILABOUR: What do you mean by data cartels?
SARAH LAMDAN: The definition of cartels varies according to differences in antitrust laws across jurisdictions, but generally, a cartel forms when manufacturers and suppliers associate in order to increase their profits and restrict competition in the markets they occupy. Data Cartels explains how major data analytics companies seem to be doing those things across information markets.
Amidst the exponential growth in data technologies and capabilities over the last 40 years (and especially with the accessibility of online access in the 1990’s) there has been a lot of consolidation in information markets, with just a few companies acquiring hundreds of publishers, data troves, and data analytics systems. These companies have taken over multiple information markets including academic and legal information markets, news and financial information markets, and also personal data markets (and especially those that are sold to major decision-makers like government entities, insurance companies, etc.).
Although the word “cartel” is often associated with illegal drug cartels, cartels are not always illegal, and they involve all sorts of goods. For example, Canada’s maple syrup cartel controls syrup prices all over the world, and in the past, cartels controlled markets for goods including lightbulbs and cheese. OPEC is a cartel that manages oil supplies to create pricing consistency in the world oil market. Illegal or not, the public should be wary of these “alliances of rivals,” because their outsized control and profit motives make them likely to engage in practices that are not aligned with the public interest. Association or “collusion” is a key part of cartel-like behavior, but those connections are hard to see. If the companies are talking or making agreements, whether implicitly or explicitly, those agreements are not usually on any public records. Ben Bagdikian, who compared the news market to a cartel, described the behavior of cartels “speaking with one voice” about their markets, even as they appear to be enemies in competition.
In the case of data analytics and information providers (including companies like RELX and Thomson Reuters), they may look like competitors, but together, they are twisting traditional content publishing into data analytics businesses, taking over all of our information markets and infusing them with data analytics products and personal data collection. They are also stifling competition in major information markets.
DIGILABOUR: How do data-savvy companies navigate across the boundaries of law/legality?
LAMDAN: In most cases, the companies benefit from a lack of applicable data laws and regulations. For example, in the U.S., Congress has yet to pass an up-to-date, comprehensive data protection law. In other cases, the companies benefit from the business-forgiving application and enforcement of existing laws. Copyright laws often give companies a wide berth to exploit information ownership, and antitrust enforcers rarely prevent information giants from taking over information markets. Lawmakers have approached data and tech companies from a place of “technological exceptionalism,” regarding digital innovation as too complex and too full of positive potential to be reined in by regulation.
This lack of intervention has allowed a few digital companies to take over the digital frontier and do, pretty much, whatever they want. A few digital companies dominate our digital infrastructure: Amazon has taken over our digital mercantile; Facebook is our public square, serving as our meeting center, news aggregator, and classified ad system; Microsoft and Google are our primary personal assistants. In this digital landscape, data analytics companies are the ones that control our digital libraries and knowledge collections, from grand digital archives to the mundane filing systems at the foundations of our business and bureaucracy, as well as the personal data dossiers that asses our criminal and social “risk.”
DIGILABOUR: Many companies sell themselves as data analytics. Does this tend to generalization? What is the role of this in current capitalism?
LAMDAN: Yes. People label all sorts of enterprises as “data analytics companies.” Some companies that people refer to as dealing in “data analytics” sell only datasets, and companies that only sell data analytics systems sans data (algorithmic, machine learning, and other forms of self-proclaimed “artificial intelligence” tools) also get called data analytics companies. A lot of companies get lumped into the generalized category of “data analytics” by reporters, scholars, and other commentators.
This is another way that tech exceptionalism provides cover for data capitalism. Without proper regulation or oversight, companies can call themselves whatever they want. Companies like Oracle and RELX can avoid explaining the full scope of their surveillance contracts by simplifying the description of their business as “data analytics” firms, and companies like Palantir can deny that they are data companies even as they build powerful predictive data systems used by police and other major decision-makers.
DIGILABOUR: What is the role of PR and lobbying in data cartels?
LAMDAN: I think the biggest way the companies’ public messaging has helped the companies flourish is keeping the full span of their business a secret. Elsevier doesn’t tell its academic journal customers that it’s part of the same company as LexisNexis, a major government data broker. Westlaw’s customers don’t know that Thomson Reuters had, and continues to have, contracts to provide U.S. Immigration and Customs Enforcement (ICE) with people’s personal data to help track immigrants. The companies have done an excellent job of obscuring the immensity of their informational power by maintaining each of their product lines in separate silos, and by obscuring what their data products do by giving them vague names like “special services” and “risk solutions.” It’s no surprise that we, as consumers, have been treating each of the companies’ market-dominating products as separate entities, and not as pieces of the same problem.
DIGILABOUR: What would an effective regulation of these companies look like?
LAMDAN: Just as there’s not a single way to peel an orange, there is not a singular form of regulation that will solve the problems caused by cartel-like data companies. We need a comprehensive set of laws and regulations to disentangle these octopus-like informational monopolies from our private laws.
One set of rules should enforce public transparency and accountability – the public deserves to know when, why, and how its data is being collected and used. The public should also be able to opt out of data collection and to fix erroneous data in these privately-owned data collections. Companies that sell and use our data should be required to make themselves and their data practices known to the public. In short, people have a right to know when their data is being collected, bought, sold, and used.
There should also be a set of rules that prevent the government from using our data without proper clearance, whether it be a warrant or some other sort of safeguard that provides people with due process and prevents people from being subject to government surveillance and data systems without a clear rationale (like probable cause) or reason (some sort of cognizable, reasonable purpose). Oher major decision-making entities that are not subject to constitutional and procedural requirements (like private insurance, healthcare, tenant and employment screening companies, and financial institutions) should be subject to similar rules that place boundaries on how they collect, use, and monetize people’s personal data.
Finally, some types of information should be publicly accessible. Scientific studies (especially government-funded ones), and legal information are sponsored by the public, and should be accessible to everyone. The information contained in academic journals, court opinions, statutes, etc. is critical for public health and safety. Companies should be fair compensated for the costs of publication and collection maintenance, but they should not be able to privatize publicly-funded information and erect exorbitantly-priced paywalls around it.
DIGILABOUR: What do you mean by digital infrastructures to support democratic ideals? How can thiese infrastructures struggle against data cartels?
LAMDAN: Among the many solutions suggested for democratizing information (aka improving public participation in our information ecosystem), the ones that I’m most optimistic about involve wresting control of public and personal information from private companies and supporting systems that give more control to everyone. We should treat information that’s created by the public, and that’s essential to public decision making, like a public resource. In contrast, we should treat private information as private, not as a resource that can be exploited by data companies.
Society could decide to fund and maintain public information infrastructures just as we maintain public roads and water systems. We could do better to oversee the companies that steward public information, and we should prevent informational oligopolies that make it harder for the public to get the information it needs and that share our private data without our consent. This isn’t the same as turning information infrastructure into a public utility – it’s about ensuring that private companies are acting in the public’s interest despite their overwhelming drive to profit from our collective informational resources. Utility and construction companies work alongside the government, and are required to provide adequate heat, roads, and clean water to everyone. Similarly, data companies should be required to provide access to critically important legal and scholarly materials, and prohibited from exploiting people’s private data without their knowledge and consent.