Automated Expulsion in the U.S. Foreclosure Epidemic

NOELLE STOUT

More than fourteen million Americans have lost their homes to foreclosure since the subprime mortgage crash of 2007, the highest rate of bank seizures in national history. Just one year into the crash, one in every 520 homes in the United States was in foreclosure. These rates were doubled in high-growth states such as Arizona, California, and Florida, where entire residential streets suddenly stood vacant. Major cities, once thriving, were forced into bankruptcy. Homelessness surged, and tent cities grew overnight. In analyzing the devastating consequences of the mortgage debacle, scholars and journalists have fingered the failures of greedy financiers, incompetent financial institutions, government agencies, and American homeowners living beyond their means. This emphasis, however, occludes an analysis of the everyday mechanics that triggered unprecedented dispossessions — in particular, how the rise of algorithmic and automated processes has shaped contemporary dispossessions.

In the wake of the crash, a dynamic interplay between automation and expulsion has emerged as calculative systems were imbued with the illusory authority of objectivity, displacing human actors and defying common sense. In this chapter, I show how the automated processes exemplified by the governmental–private-sector loan modification bureaucracies established in 2009 and the robosigning scandal emerging in 2010 systematized and standardized the foreclosure process. New systems designed to maximize efficiency and thus profits for investors often introduced rampant errors, prevented humane decision making, and punished those homeowners who most qualified for mortgage assistance. To illustrate how automated processes have led to unparalleled numbers of bank seizures of American homes — in the millions — I draw on long-term research beginning in 2012 among homeowners applying for loan modifications and the lending employees processing their appeals in California's Sacramento Valley, one of the hardest-hit regions in the nation.

Here I focus on the automated systems that have enabled millions of foreclosures, but it is important to recognize how similar algorithmic processes played a significant role in the expansion of high-risk subprime lending and the Wall Street financialization of U.S. mortgage markets that led to the 2007 mortgage crash. Wall Street investment firms used technological innovations to relax underwriting standards that determined a mortgagor's qualifications. Within months after a new homeowner signed a subprime mortgage contract, mortgage loans were bundled into collateralized debt obligations and sold as securities in a secondary derivatives market, which relies on algorithmic technologies to trade in contracts, such as futures and options, based on other assets. The pools of mortgages would be divided into tranches, often organized around their risk level. Through this reselling, high-risk subprime loans were mislabeled as safe investments and enmeshed in high-value stocks in retirement and other investment funds. At each step in the process, investors and loan officers depended on algorithms and automated systems to transform mortgagors and their long-term debts into abstract payment streams.

In a system where incentives depended on commissions and transactions, lenders and mortgage brokers could take advantage of so-called creative mortgage products — technologically driven formulas that produced mortgage loans — while deferring the decision making, and the responsibility, to automated systems. Pressured to keep up with Wall Street's demand for mortgage loans to feed secondary markets, many brokers and lending employees pushed high-risk, high-interest loans on mortgagors who would ordinarily not qualify for them. Underwriting algorithms allowed borrowers to qualify against their best interests, such as elderly couples living on fixed incomes who were convinced to take a second mortgage on a home that was already paid off. Other homebuyers who qualified for conservative, fixed-rate mortgages with lower interest rates were duped into taking high-risk loans with introductory teaser rates that would balloon within a matter of years because commissions were higher for these subprime products. Meanwhile, mortgage brokers and lenders exposed themselves to little risk: within months the mortgage loan would be securitized and sold on a booming derivatives market. What's more, when the scheme collapsed, Wall Street investors correctly surmised that a government bailout would rescue them.

Whereas before the 2007 collapse, automated underwriting standards were relaxed to saddle homeowners with loans they could not afford, after the crash, different algorithmic formulas forced millions of homeowners seeking to modify their loans to default. At first glance, these foreclosures might seem straightforward: a homeowner falls behind on mortgage payments and a lender seizes the home, which served as collateral for the loan. But in the aftermath of the crash, lenders and loan servicers employed algorithms to determine whether to proceed with evictions and relied on automated protocols to carry out bank seizures that are anything but commonsensical. According to many bank employees adjudicating these processes, before the use of these automated systems, employees had more leeway to decide the worthiness of homeowners seeking assistance. Lauded for their ability to limit human error and increase efficiency, these automated practices have triggered the dispossession of millions of Americans, embroiling mortgagors in byzantine dramas leading to illogical outcomes and ethically suspect home seizures.

The Maze of Loan Modification

With plans to expand their family, Susan and Rick Condit, a white married couple in their late twenties, purchased a modest three-bedroom home in a lower-middle-class Sacramento suburb for $250,000 in 2006, at the height of the market. Their monthly mortgage was a stretch, but they cut back on expenses to cover their payments. By 2009, as mortgage markets collapsed, their home was worth less than half of the amount they paid, with no hope of recovering its value. Due to severe California state cutbacks during the Great Recession, Susan lost her job as a first-grade teacher in the public schools. Rick, who had made a decent living installing air-conditioning units in new houses during the construction boom, found work drying up as the mortgage crash took its...