With her expertise in both administrative and criminal law, Rachel Barkow has a unique perspective on the criminal justice system. She has analyzed sentencing commissions (she currently sits on the US Sentencing Commission), prosecutor’s offices, and, most recently, the structure for deciding executive clemency, through her administrative law lens.
In her inaugural Segal Family Professorship of Regulatory Law and Policy lecture last November, Barkow examined the utility of her framework, arguing that while the criminal justice system is a regulatory one, it does not incorporate the same checks on its power as other regulatory systems—to criminal justice’s detriment. “The idea would be to start using data to make decisions, as opposed to just people’s gut instinct,” says Barkow in an interview. “That could apply across a range of criminal justice decision-making points, from prosecution to sentencing to policing to clemency.”
Barkow pointed to an absolutism in the regulation of criminal behavior that sets it apart from other regulatory realms. “It’s essentially zero tolerance for any risk,” she said in her lecture. “One story, and politicians are willing to take an entire program down without considering whether the program, on net, brings more benefits than it has costs and whether it reduces risks overall…. We don’t approach any other area of government regulation this way,” she said, referring to, say, vaccines, environmental policy, and financial regulation.
Willie Horton, a convicted murderer in Massachusetts who escaped while on weekend furlough to rape a Maryland woman and beat her spouse, is an example of how one awful story can derail a government program without closer analysis of whether the risks of the program are outweighed by the benefits it brings. George H. W. Bush successfully invoked Horton during the 1988 presidential campaign to attack Massachusetts Governor Michael Dukakis as being soft on crime. Since then, programs have not been rationally assessed in terms of costs and benefits but are discarded if they pose any risk to politicians for Horton-like stories.
The focus on harsher criminal punishments in recent decades originated in concern over rising violent crime rates in the 1960s and 1970s. Barkow argues, however, that even as those numbers stabilized, the crackdown expanded to a wider range of crimes and criminals: “The question is, do long sentences make sense for nonviolent offenders?”
With an administrative law scholar’s attention to the structure and creation of an agency, Barkow thinks about how to design a criminal justice agency to avoid conflicts of interest and cognitive biases, while also considering all stakeholders. For example, crime victims are critical stakeholders, Barkow says, but asserts that “the way we usually address their needs is to pass some symbolic legislation that actually doesn’t help them at all.” Enhancing safety to create as few victims as possible is the best first step, she says, along with bolstering resources to help victims restore their own lives. For instance, when New Mexico abolished its death penalty, it redirected the savings to a victim restitution fund.
Different kinds of criminal justice reforms could call for different solutions, she adds. In the case of clemency, the focus of her current work, an advisory agency in the Office of the White House Counsel or somewhere in the Office of the President, rather than in the Department of Justice, would help avoid the conflict-of-interest issues inherent in having the DOJ both prosecute individuals and later weigh in on their pardons, as is now the case.
Barkow notes growing bipartisan support for less draconian, more data-based criminal justice approaches, including the Fair Sentencing Act in 2010, which reduced the disparity in sentencing for possession of crack versus powder cocaine, and a greater emphasis on cost-benefit analysis like that performed by the Washington State Institute for Public Policy.
“At the end of the day,” says Barkow, “government is supposed to keep people safe and solve problems. Some of these things backfire and don’t promote public safety, and they’re expensive to boot.” She says she is not coddling criminals, but instead asking a more practical question: How do you promote public safety in the most efficient, data-supported way? “That is a political debate that can be had without costing people elections,” Barkow says. “In fact, it can win people elections.”
Posted June 20, 2014