This essay is part of a symposium on the prospects and dangers for human rights in the governance of digital platforms. Read the rest of the series here. New posts will be published between June 18th and June 30th 2024.
Joahna Kuiper / Better Images for AI / Small Data House (square) / CC-BY 4.0
In many parts of Canada and the United States, housing markets have become what philosopher Debra Satz calls toxic markets. A market is toxic if it is morally concerning or harmful because of the types of goods and services traded, the conditions under which the transactions take place, or the outcomes of such transactions. In the case of housing, market toxicities are manifestations of growing inequality in the power relationship between renters and landlords, the difficulty of accessing homeownership for first-time homebuyers, high rates of homelessness, and supply shortages. To address the problems surrounding housing markets, local and national governments have proposed numerous policy solutions. However, little attention has been paid to the socio-technical development that has created many of the conditions that exacerbate the housing market’s toxic effects: datafication.
Datafication occurs when a subject is monitored and recorded. The perceived facts that are recorded are what we call data, an abstraction of the material world. Housing data is the recorded information that is generated from the monitoring of the housing ecosystem. This data is not inherently bad, as it can be leveraged for many positive purposes, such as better distribution of community resources. However, in recent years, the general trend of increased datafication due to the Internet revolution has led to housing data being leveraged to generate economic benefits rather than to build stronger communities. Thus, the increase in housing data has created a bifurcation of the housing market. Traditionally, from the buyer’s perspective, the housing market is primarily a means to acquire housing. We argue that the increase in datafication is creating and reinforcing a change in the market. As market practices shift towards a financialized conception of the housing market, housing has become increasingly harmful for people looking for accommodation.
To become a financial product, an object must be calculable, i.e., calculable into an economic formula. This is achieved through datafication. Research conducted by Megan Nethercote in Australia focuses specifically on the role of home appraisers in the financialization of the Australian housing market. Appraisers convert the physical structure and social status (i.e., neighborhood) of a home into a calculable economic value. It has become much easier to assess the value of a home, especially in recent years, as data on home sales and values has become more readily available and home values can be easily estimated using algorithms rather than trained human appraisers. As Don Layton, former CEO of Fannie Mae, pointed out, the digitization and democratization of housing data through easier access to housing data and lending information has allowed individual market actors to visualize their home as a financial product. When people can visualize their home as a financial asset and obtain a valuation without friction, they are conditioned to treat their home like a financial product, becoming more aware of value appreciation and financial options such as mortgages. Treating housing as merely a commodity leads individuals (and investors) to act in a way that maximizes their return on their investment, rather than focusing on the well-being of the individuals who live in the house or apartment.
The financialization of housing, driven by increasing information on house prices, has led to the phenomenon of housing platformization. Platformization is taking place in two main ways: first, the platformization of the house itself, and second, the platformization of the housing market. The platformization of housing is seen as a growing trend towards renting over traditional home ownership in the Global North. Housing stock is increasingly owned by non-tenant agents, who then become contractual clients of the owner. This effect is amplified by the previously discussed ease of mortgaging a home, made possible by the increased availability of housing data. It is now common practice to leverage investment properties to expand your portfolio.
The home rental market itself has also been platformized by services such as Airbnb. Airbnb is a good example of how certain uses of platforms can have a negative impact on the housing market and on relationship equality. The housing market has been further platformized through the development of platforms that act as real estate market intermediaries. Companies such as Zillow and realtors.com act as digital listing agents that connect home sellers and buyers. Through this process, these platforms are able to accumulate vast amounts of information, through the listings posted on their platforms, about the market itself as well as other market factors derived from the analysis of user behavior and actions. This creates an unequal market as one side has a significant information advantage.
Recently, in the housing platform space, a related phenomenon called iBuying has emerged. iBuying refers to companies that act as true market intermediaries. These companies collect data about the housing market and use it to fuel algorithms that assess the value of homes and predict their future value. As such, these companies purchase homes with the intent of reselling them for a profit. One of the leaders in this space is US-based Opendoor. Opendoor offers convenience to homebuyers by instantly providing a fast and “painless” cash offer for a seller’s home based on the information the seller provides. The offer amount is determined by an algorithm based on Opendoor’s predictions about the property’s resale potential. iBuying’s model is very similar to the virtual stockbroker model (i.e. Robinhood) that aims to make it easier for the average person to trade financial assets on the stock market. This is a powerful illustration of how the proliferation of housing data is creating an environment in which homes are increasingly treated like financial assets, thereby limiting the potential for individuals to interact on an equal footing within (and around) that market.
It is our view that while governments recognise the negative impacts of platforms like Airbnb, they have not adequately addressed the underlying cause: the datatification of housing. Moreover, current legislative/regulatory approaches to address the social harms caused by datatification are not adequate to address harms such as the disregard for tenants’ rights and, to some extent, gentrification seen in the housing market. Specifically, the two main approaches to data regulation are currently algorithmic regulation and privacy regulation. The latter is represented by laws such as the General Data Protection Regulation (GDPR) in Europe, but as it focuses on protecting citizens from the misuse of their personal data, it does little to address the type of housing data we have examined in this article. This is why algorithmic law is essential; as it can be a powerful tool to ensure that algorithms run on housing data do not exacerbate certain social problems, such as race-based discrimination in housing. However, such laws do little to address the social impacts that arise from the intended application of the algorithms themselves. Algorithms such as those used by iBuying companies, even if they seek to act in a completely fair manner, have a negative impact on the market and undermine the social objective of providing housing.
While we cannot yet offer concrete solutions, we draw attention to the fact that broad data protection policies are insufficient to protect citizens from the harms of datafication. Datafication of different subjects produces different consequences. The datafication of housing reinforces the treatment of housing as a financial product, thus reshaping the housing market and making it harmful for citizens looking for housing. We therefore believe that data policies must also adopt a targeted approach and adopt customized solutions that address the individual social harms produced by the datafication phenomenon. In addition to measures that favor non-profit housing and provide stronger protection for tenants, other solutions must focus on the root causes of the housing crisis, such as wealth concentration, rent-seeking, and ultimately the idea of homeownership as the solution to all problems.
