Virginia Eubanks
In the fall of 2008, after Indiana launched its automated benefits eligibility system, Omega Young was notified that she needed to recertify for Medicaid. Young suffered from cancer, and on the day of her appointment she was undergoing treatment and unable to attend. She placed a call to her local help center to let them know. Nevertheless, the system registered that she “failed to cooperate,” and cut off all her public assistance, including food stamps, healthcare, and pharmaceuticals.
Young embarked upon a months-long appeal to restore her benefits, during which she was unable to afford medications and struggled to pay her rent. Finally, on March 2, 2009, Young got her benefits back. She had died on March 1.
Automating Inequality, Virginia Eubanks’ new book about how automated eligibility systems “profile, police, and punish” the poor, is loaded with horror stories like Young’s. They are not all as tragic, but serve as alarming evidence that Americans continue to treat poor people as second-class citizens. Eubanks’ central premise is that the poor are largely portrayed, particularly by conservatives, as either criminals or freeloaders and as the main problem with American society. Because of this portrayal, Americans tolerate systems that dehumanize and surveil the poor to a degree that would not be tolerated if the systems were designed for other classes.
Americans tolerate systems that dehumanize the poor to a degree that would not be tolerated if the systems were designed for other classes.
While supporters of automated systems claim they increase efficiency and remove prejudice from decision making, Eubanks contends that, in fact, these systems possess inherent biases and make benefits harder to obtain, overturning the gains won by the welfare movements of the 1960s and 1970s.
Eubanks’ argument is powerful, but the book would have benefitted from a deeper discussion of income inequality and the policies that aggravate it. Indeed, the gap between rich and poor in America is wider than it’s been since the Great Depression, and it’s harder than ever for poor Americans to ascend the income ladder. Policies such as trickle-down economics, mass incarceration, and outsized court fees help keep the poor poor. Automated eligibility systems, Eubanks reveals, also play a role.
The biggest victory of the welfare rights movement, Eubanks tells us, was a court ruling that redefined welfare as personal property, not charity. This meant due process had to be provided to recipients before benefits were removed. The ruling led to a long stretch when the poor could easily access benefits.
But in the 1980s, conservative politicians began portraying recipients as lazy blacks, and the media ran with it. Restrictive new rules were implemented and hyperbolic stories of rampant welfare fraud appeared in newspapers and on television. All this led to what Eubanks calls a “political sleight of hand”: politicians claimed automated systems reduced fraud and increased efficiency while the real intent was to degrade due process and shrink welfare rolls.
She offers these eye-opening statistics: In 1973, nearly half of Americans living below the poverty line received much-needed AFDC benefits. Today, that number is less than 10 percent. Also in 1973, four of five children living in poverty received benefits. Today—in our far, far richer nation—fewer than one in five poor kids do. Automated systems, Eubanks writes, “acted like walls, standing between poor people and their legal rights.”
Indiana’s automated system was launched because Republican governor Mitch Daniels, a long-time foe of public assistance, saw that his state was slow at moving citizens from welfare to work. Although the existing system was flawed, Eubanks relates, it offered recipients easy access to caseworkers.
The new system, designed by corporations including IBM, largely removed human decision making, and kicked slews of eligible Hoosiers off the welfare rolls. It made so many mistakes, in fact, that some politicians believed it was purposely designed to discourage the needy from claiming benefits. During the first three years of automation in Indiana, the error rate tripled, and most errors were negative: They wrongly stopped people on food stamps from accessing them.
Indiana was forced to ditch its system after a public outcry. But Eubanks floats a provocative question: Did it really fail? Sure, the system was an unmitigated disaster for Omega Young and thousands of others, but IBM walked away with millions and Indiana drastically reduced its welfare rolls. Wasn’t that the goal?
Indeed, an issue I would have liked to see further explored in Automating Inequality is profiteering; how these privatized systems that deprive the poor of benefits make vast sums for the corporations that design them. Eubanks touches on this when she dubs automated systems the “digital poorhouse” and compares them to the poorhouses of the 1800s, where owners exploited inmates’ labor to run side businesses. Beyond the Indiana-IBM example, Eubanks doesn’t spend much time on it.
Digital surveillance and predictive risk models, on the other hand, are thoroughly explored. Turning to L.A.’s “coordinated entry system,” Eubanks reveals how data mining is used to restrict the poors’ rights. She describes L.A.’s system, which surveys Skid Row’s homeless to prioritize them for housing, as “a surveillance system for sorting and criminalizing the poor.”
Eubanks reveals how data mining is used to restrict the poors’ rights. She describes L.A.’s system, which surveys Skid Row’s homeless to prioritize them for housing, as “a surveillance system for sorting and criminalizing the poor.”
Eubanks notes that data collection is only as good as the collector, so bias and inaccuracies are an inherent part of the process. In interviews, the homeless are encouraged to be forthcoming, but when they admit to illegal behavior they risk police scrutiny. Easy law enforcement access to this data “threatens to turn routine survival strategies of those living in extreme poverty”—sleeping in the wrong spots, sitting on sidewalks, panhandling—“into crimes.”
For her final study, Eubanks examines a risk model in Allegheny County, Pennsylvania, aimed at preventing child abuse and neglect. To make a prediction, the system sifts a trove of data on parents’ interactions with the county’s welfare, child protective, and criminal justice systems.
But as Eubanks points out, “nearly all the indicators of child neglect are also indicators of poverty.” A child of a poor single parent, for example, is more likely to spend time alone, to live in a less-safe neighborhood, and to live in a cluttered house (all risk factors). But these hardly add up to abuse or neglect.
Another major problem with the model is that all the data comes from the public employees who interact with poor parents and make judgments about their lives. Data about middle-class families—who use private sources of family support—is not included.
“Imagine the response,” Eubanks notes, “if Allegheny County proposed including data from nannies, babysitters, private therapists, Alcoholics Anonymous, and luxury rehabilitation centers to predict child abuse among wealthier families.”
While the county says it hopes to include private insurance data to make the pool more representative, it admits this won’t happen anytime soon. “The professional middle class will not stand for such intrusive data gathering,” Eubanks says, paraphrasing a county employee.
In May 2010, the state of Indiana sued IBM for $437 million, blaming the company for all the faulty benefit denials that wreaked so much havoc in the lives of its poorest citizens. IBM countersued for $100 million. In his decision, the judge said neither party deserved to win. “This story represents a ‘perfect storm’ of misguided government policy and overzealous corporate ambition,” he wrote. “Overall, both parties are to blame and Indiana’s taxpayers are left as apparent losers.” But the biggest losers weren’t necessarily taxpayers.
“In the end,” Eubanks observes, “the Indiana automation experiment was a form of digital diversion for poor and working Americans. It denied them benefits, due process, dignity and life itself.”
Jake Whitney is a journalist based in New York.