Why I Still Believe in Sentencing Guidelines (Just Not the Current Federal Guidelines)
I am thrilled to join Professors Berman and Chanenson in this new platform for commentary on sentencing law, policy, and practice. And I’m grateful to them for allowing me to join the wonderful corps of editors at the Federal Sentencing Reporter (FSR).
I had the great pleasure of spending time years ago with Professor Dan Freed, a founder of FSR, and visiting his sentencing program at Yale Law School. Freed was a giant in the world of sentencing, but more importantly, he was a gentle, soft-spoken, brilliant, and wise teacher. His combination of passion for sentencing and warmth towards others made him a uniquely effective educator. It was both a joy and a great learning experience to be in a classroom with him and with his colleagues like Professors Marc Miller, Kate Stith, and Nancy Gertner.
I have published many articles in FSR over the last 30 years or so, and it is now a genuine honor to be included among the editors of the journal Freed and Miller started so many years ago. I’m very much looking forward to editing issues in the coming years. And I’m grateful for the chance, in this forum here on Substack, to share some thoughts now and then on sentencing law, policy, and practice – and justice too.
- - -
Like many before me, I first fell into the rabbit hole that is federal sentencing law and policy as a practitioner. For me, it was in the late 1980s. I’ve been stuck in that hole ever since. Over the last thirty years, after serving as an assistant public defender and a federal prosecutor, I was on staff at the federal Sentencing Commission, but mostly part of the criminal justice policy apparatus at the U.S. Department of Justice across many administrations. For ten years, ending just about six weeks ago in May, I represented the Attorney General on the Commission. I was privileged to work with all the federal Commission Chairs: Judges Wilkins, Conaboy, Murphy, Sessions, Hinojosa, Pryor, Saris, Breyer, and Reeves. A raucous roller coaster ride indeed.
In addition to being a regular reader of FSR, I’ve been a follower of Professor Berman’s blog since it first appeared just as the Supreme Court was putting its finishing touches on Booker almost 20 years ago. Like FSR, the blog has been an indispensable companion to those of us who continue to grapple with the unanswerable questions surrounding the imposition of state-sponsored punishment on those who do wrong. Thank you, Professor, for the work you have put into keeping the blog going these many years.
I will have plenty to say in coming posts about the various Commissions and commission actions over the last 30 years. But I thought I would start with a few words about why I still believe in, and support the idea of sentencing guidelines, but not the federal sentencing guidelines as they are currently crafted.
In late March, Professor Daniel Kahneman, a Nobel laureate and a great psychologist and economist, died at age 90. I remember vividly, Commission Chair Patty Saris walking into a Commission meeting in the early 2010s carrying Kahneman’s best-selling book, Thinking Fast and Slow. She raved about the book and its insights on rationality, irrationality, and human decision-making generally; about how it exposed the foibles of the way we, flawed humans all of us, go about making choices in our professional and personal lives; and about the framework Kahneman lays out – thinking fast and slow – to understand how psychologically we go about our days and the choices we face.
I was first exposed to some of the psychological scholarship underlying Kahneman’s work – and how it applies to sentencing decisions – when I spent time at the Sentencing Commission in the 1990s with the late Professor Richard “Barry” Ruback. Barry was a Supreme Court fellow at the Commission when I was Deputy General Counsel there and when Judge Richard Conaboy began a project – a failed project would be a fair assessment – to simplify the guidelines. Barry, like Kahneman, was generous and giving and thoughtful. He would spend much of his career as a Professor of Criminology and Sociology at Penn State University. Barry knew psychology and taught me and others about heuristics and biases, noise and overconfidence, probabilities and risks, the frailties of human decision making, and how those frailties manifested themselves in judges imposing criminal sentences.
Barry and I would later write an article together, “The Federal Sentencing Guidelines: Psychological and Policy Reasons for Simplification.” It was published in 2001 in Psychology, Public Policy, and Law. It holds up pretty well today, all these years later, mostly because the psychological scholarship of Ruback, Kahneman, and many others – and their insights on human decision making, which are the foundation for the article – holds up well. And in marking the death of both Ruback and Kahneman, we ought to recognize their great contributions, including understanding how structured decision making – especially in a complex endeavor like sentencing, with conflicting goals, high stakes, and many factors to consider – will lead to more consistent, proportional, and goal driven results than unstructured decision making.
The insights of these scholars also reveal that while a structured and actuarial model of decision making will outperform most humans on consistency and along other dimensions, greater detail in the decision-making structure, like the federal Commission decided to incorporate in drafting the federal sentencing guidelines, does not necessarily bring with it greater validity across dimensions. In fact, much research suggests that attempts at greater precision will often lead to lower validity. This is the experience of the federal guidelines. The guidelines were drafted to dictate precise sentencing outcomes, and many judges recoiled. But more than that, the attempts at precision ran into psychological and constitutional limitations. Especially now, 20 years after the Booker decision, it is time to revisit the guideline structure created by the first Commission.
The guidelines are a case study in a failed attempt to use brute force – and too much precision – to get to particular algorithmic outcomes. Of course, there is another way to bring consistency and sound policy to sentencing decisions. As Richard Thaler and Cass Sunstein suggested in their best-selling book, Nudge, which was also drawn from Kahneman’s work, policy makers can create a less dictatorial choice architecture to try to move decision makers, like sentencing judges, in a particular direction. With psychological scholarship in mind, it is not difficult to imagine a simpler federal sentencing decision tree, one ironically with even greater richness than the current federal guidelines and yet simultaneously more understandable to all, including those being sentenced, the victims of crime, and the general public. One need only look to guidelines system of many states and certain foreign countries that have learned from the U.S. federal system what not to do. Most everyone familiar with the federal guidelines can agree upon this: they are not nuanced; they are not a nudge; and they are not a model.
I’ve written a lot – for the Department and on my own – about the structure of the federal sentencing guidelines and how it might be reformed to lead to better sentencing outcomes. I will have more to say in future posts. But for now, I just want to remember Kahneman and Ruback and all of their psychologist and behavioral economist colleagues who shared insights that should drive the architecture of sentencing decision making and the use of sentencing guidelines.
More to come.