A few days ago, the invaluable Liz Oyer noted a change in the Bureau of Prisons (BOP) rules. That agency runs our federal prisons, directing the lives of over 150,000 federal prisoners. It is a change in the rules that seems innocuous until you start thinking about it. Briefly, the old rule was that the BOP decides which prison each inmate goes to; the new rule is that the nation’s attorney general can override the BOP’s decision.
Here is the way things used to be: The BOP conducts a detailed, bureaucratized, quantitative assessment of each inmate’s “security designation and custody classification.” Each inmate is assigned a score based on factors such as offense severity, criminal history, sentence length, and age. That score is used to determine where the inmate serves time.
This is a high-stakes assessment because some federal prisons are much more unpleasant places to endure than others. The nature of the prison largely determines the quality of the inmate’s life. Prisons are on a spectrum from minimum security (no cells, bars, fences, or armed towers; dormitory-style housing; significant freedom of movement for prisoners around the facility; violence is unusual) to high security (lethal electric fencing; much cell time; heavily regimented prisoner movement that typically requires an escort; a high percentage of dangerous inmates).
The BOP modified the rules of prison assignment in May. The new rule allows the nation’s attorney general to override the results of any inmate’s quantitative security designation. The new rule (an amendment to BOP’s Program Statement 5100.08) allows the attorney general, personally, “to designate or redesignate the place of a prisoner’s imprisonment” at his discretion. The attorney general can now direct the BOP to transfer any particular inmate to any particular prison, based on the attorney general’s exclusive judgment.
Does it matter how inmate placement is decided? As a practical matter, is it important whether a sub-Cabinet agency makes the call on such decisions, or whether the attorney general can supersede BOP’s decisions? I think it’s very important.
Put it this way: The old rule appeared to have several qualities that the new rule lacks. The decisions that came from the old rule rested on a formalized, bureaucratic process that attempted to treat inmates neutrally and equally; that made decisions on the merits; that de-emphasized calibrations of politics, punishment, and vengeance and emphasized orderly administration and public safety; and that rested on a decisional process that used local information to make non-centralized decisions. (At the very least, the old rule appeared to insulate itself from a certain kind of politics.)
Keeping those decisions out of the hands of the nation’s top prosecutor was a management choice with benefits that may not have been immediately apparent. One of those benefits is the distinction it highlighted between the prosecutor’s central responsibility (that is, to convince the court of the defendant’s guilt) and the judge’s and jury’s responsibility (that is, to ultimately determine guilt and assign punishment).
Even before this rule was changed, it was under pressure. Under the Trump administration, the death-row federal prisoners whose sentences had been commuted to life without parole by President Biden were treated as harshly as possible, per a Trump executive order that demanded that the offenders be “imprisoned in conditions consistent with the monstrosity of their crimes.” Many were transferred to ADX Florence, the most dangerous, unpleasant, and restrictive penitentiary in the nation. This executive order replaced one value with another—it replaced a focus on the orderly yet decentralized administration of prisons with a focus on calibrating individual punishment. (And, again: the calibration of individual punishment is a matter that should be decided by a judge and a jury at least as much as by a prosecutor.)
The pressure on the BOP rule continued throughout the Trump administration, as demonstrated by the transfer of Ghislaine Maxwell to a minimum-security prison. (Maxwell was convicted of the sex trafficking of young girls for the benefit of the notorious Jeffrey Epstein.) Her transfer cannot be reconciled with the contemporaneous demands of Program Statement 5100.08, which required sex offenders to be placed in relatively high-security facilities in order to protect the surrounding community. That transfer occurred just a few days after Todd Blanche, then the deputy attorney general, met with Maxwell for approximately nine hours over two days.
It was a meeting that furthered the political interests of the president: During those meetings, Maxwell affirmed that President Trump had been “a gentleman in all respects” and had “absolutely never” behaved inappropriately when she had been in his company and Epstein’s. Just a few days later, Maxwell was transferred to a minimum-security prison, which she has described as safer, more comfortable, calmer, cleaner, and less violent, with more polite staff and better food. The BOP’s point system does not allow a sex offender to be transferred to a minimum-security facility—one that ordinarily houses non-violent offenders and white-collar criminals—absent special circumstances, and it looks as if the special circumstances here are nothing more than executive fiat.
After Maxwell’s transfer, the BOP issued a defensive statement claiming that Maxwell’s transfer was based chiefly on inmate safety concerns, that the decision was made by the BOP, that the new facility offered improved security measures, and that no preference or political influence played any role in the transfer decision. The BOP’s statement is not credible. Inmate safety is simply not a factor that is part of the calculation of the BOP score. However, public safety is and there is no sense in which public safety is furthered or in which the BOP’s rules are honored by transferring a sex offender to a lower-security facility. Rather, inmate safety concerns—the concerns that the BOP’s statement cloaked itself in—are typically addressed by removing an inmate from the general population to more secure protective custody within the same facility.
Almost all BOP prisons have such secure sub-facilities; inmate-safety concerns are a major reason those sub-facilities exist, and addressing an inmate-safety concern with an external transfer is an expensive and highly unusual departure from administrative norms. It seems clear that Maxwell didn’t want more security; she wanted more freedom and more privileges, and she got them. There is no sense in which Maxwell’s new situation, a minimum-security facility, offers higher security; it offers a lower inmate-to-guard ratio and more personal freedom for inmates. The notion that the transfer decision was made by BOP and not by Blanche is difficult to take seriously; it stretches credulity to suggest that the timing of the transfer, just after the attorney general’s two-day meeting with Maxwell, was merely a coincidence.
Again, the new rule allows the attorney general “to designate or redesignate the place of a prisoner’s imprisonment” at his discretion. As shown above, it appears that, in practical terms, the old rule had already been breached. In my opinion, the circumstances of that breach make for a persuasive argument that the old rule is far superior to the new one.
In the domain of particular applications of the law to particular instances of criminal justice, I think it’s a good idea to preserve some separation between the decisions of the BOP and the decisions of the attorney general—for reasons that are roughly analogous to the traditional separation between the White House and the Department of Justice. If decision-making processes can be neutral, they are a better locus for administrative enforcement decisions in individual cases than political actors; bureaucratic processes are more likely to focus on public-interest concerns and less likely to be distracted by political interests. The notion that the circumstances of imprisonment decisions should rest on anything but the public interest (primarily as reflected in the orderly administration of prison facilities and the safety of the surrounding community) is difficult to defend; in fact, a federal statute prohibits favoritism in prison assignment for “prisoners of high social or economic status.”
When Oyer announced the BOP’s rule change, she suggested that this could open the door to a president who asks his attorney general to assign political opponents to maximum-security prisons while assigning political allies to significantly less unpleasant prisons. This nightmarish outcome—an American public official taking the political views of convicts into account when making sentencing decisions—might be seen as implausible by some people; it appears less implausible when we recall that Ed Martin, one of Trump’s pardon attorneys, was famously quoted as saying, “No MAGA left behind.”
In Slaughter, the Supreme Court provided extensive grounding for the general rule that executive-branch decisions should be made by those at the top of the executive branch and not delegated to others. But there is a prudential exception to that general rule. That exception involves respect for a norm of decentralized, non-politicized decision-making in certain realms. Those realms certainly include some kinds of particularized criminal justice decisions. These are decisions that are better made in a decentralized manner by administrators, and keeping them decentralized will yield payoffs in both public confidence in the integrity of administration and insulation from vengeance and favoritism. A prudent chief executive will delegate such decisions to subordinates, and a new rule that centralizes decision-making authority by pushing it up the chain of command is a form of institutionalized imprudence.
