Why Expanding Compassionate Release Is a Dangerous Departure From Sentencing Certainty

The Sentencing Reform Act's Promise of Finality—Now Under Siege

In my 25 years as a federal prosecutor and now as a defense attorney, I have watched the federal sentencing system evolve from a coherent, determinate structure into something increasingly ad hoc. The Sentencing Reform Act of 1984, codified at 18 U.S.C. § 3551 et seq., was Congress's answer to the indefensible disparities of indeterminate sentencing. It created a framework where sentences were meant to be certain, proportional, and final—subject only to narrow, statutorily defined exceptions. Chief among those exceptions was compassionate release, found at 18 U.S.C. § 3582(c)(1)(A), which permitted a sentencing court to modify a term of imprisonment only upon a finding of "extraordinary and compelling reasons." For decades, this provision was applied sparingly, reserved for inmates with terminal cancer, debilitating neurodegenerative diseases, or extreme age combined with serious infirmity. The Bureau of Prisons held the gatekeeping power, and motions were rare precisely because the standard was so demanding. That gate has now been thrown wide open, and I can tell you from direct experience that the consequences for sentencing certainty are catastrophic.

The expansion began in earnest with the First Step Act of 2018, which amended § 3582(c)(1)(A) to allow inmates to petition courts directly after exhausting administrative remedies. Suddenly, the gatekeeper role shifted from the BOP to the judiciary, and the volume of compassionate release motions exploded. By 2020, federal courts were receiving tens of thousands of such motions annually, many of them based on COVID-19 exposure fears—a circumstance Congress never contemplated as "extraordinary" within the meaning of the statute. The U.S. Sentencing Commission responded by promulgating emergency guidelines under U.S.S.G. § 1B1.13, which attempted to define "extraordinary and compelling reasons" in broader terms. But these guidelines went far beyond the original statutory intent. They included categories for family circumstances, victim abuse, and even "other reasons" that essentially invited judges to create their own grounds for release. In my practice, I now see defense attorneys filing compassionate release motions for clients who are simply unhappy with their guideline range—a distortion that would have been laughed out of court a decade ago.

This erosion of finality strikes at the heart of the Sentencing Reform Act's bargain. Defendants who plead guilty do so in reliance on a predictable sentence; prosecutors extend offers based on that same predictability. When compassionate release becomes a routine backdoor to resentencing, that bargain collapses. I have personally handled cases where a client rejected a favorable plea agreement because they believed they could "win at trial and then get compassionate release later if things go wrong." That is not a system of justice—it is a system of gamesmanship. The statute's plain language requires "extraordinary and compelling reasons," not mere dissatisfaction with the original sentence or changed personal circumstances. Congress knew what it was doing when it used those words, and courts that stretch them into pretzels are rewriting legislation from the bench.

How Judicial Expansion of U.S.S.G. § 1B1.13 Creates a Shadow Sentencing System

The U.S. Sentencing Commission's 2023 amendments to U.S.S.G. § 1B1.13 represent the most aggressive expansion of compassionate release authority in federal history. Under the revised guideline, courts are now permitted to consider a non-exhaustive list of "extraordinary and compelling reasons" that includes changes in the law—even non-retroactive changes—as a basis for reducing a sentence. Let me be blunt: this is a direct end-run around the constitutional prohibition on ex post facto applications of criminal law and Congress's explicit decision to make certain guideline amendments retroactive only when specifically designated. The Sentencing Reform Act deliberately limited retroactive application of guideline changes to those listed in U.S.S.G. § 1B1.10, and for good reason—finality matters. When a defendant was sentenced under a 100-to-1 crack-to-powder ratio in 2005, that sentence was lawful and final. The fact that Congress later reduced that ratio through the Fair Sentencing Act of 2010 does not make the original sentence unjust; it simply reflects a policy change that was not retroactive.

Yet I now see federal judges in the Second, Ninth, and D.C. Circuits routinely citing non-retroactive guideline amendments as "extraordinary and compelling reasons" under § 1B1.13. In one case I consulted on, a defendant sentenced in 2012 for a career offender enhancement under U.S.S.G. § 4B1.1—which was later modified by Amendment 798—successfully argued that the change itself constituted an extraordinary circumstance, even though the Sentencing Commission explicitly declined to make Amendment 798 retroactive. The court granted the motion, effectively nullifying Congress's decision and the Commission's judgment. This is not compassionate release; this is judicial legislation. The statute at 18 U.S.C. § 3582(c)(1)(A) requires the court to consider the factors set forth in § 3553(a), including the need for the sentence imposed to reflect the seriousness of the offense and provide just punishment. Granting release based on a non-retroactive guideline change undermines every one of those factors.

The practical effect of this expansion is a chaotic patchwork of outcomes that depends entirely on where an inmate is incarcerated. A defendant in the Southern District of New York might receive compassionate release for a non-retroactive guideline change, while an identical defendant in the Eastern District of Texas is denied. This forum-shopping destroys the uniformity that the Sentencing Reform Act was designed to achieve. Furthermore, the expansion encourages frivolous litigation. I have reviewed compassionate release motions that cite everything from weight gain to marital discord as "extraordinary and compelling reasons." Each of these motions consumes judicial resources, delays meritorious cases, and erodes public confidence in the finality of criminal sentences. The Commission's guidance under § 1B1.13, by including a catch-all "other reasons" provision, has given litigants a blank check to argue almost anything—and too many courts are cashing it.

The Dangerous Precedent of "Rehabilitation Alone" as a Basis for Release

One of the most troubling developments in compassionate release jurisprudence is the growing acceptance of rehabilitation alone as an "extraordinary and compelling reason" for sentence reduction. The First Step Act explicitly provides at 18 U.S.C. § 3624(b) that rehabilitation efforts shall not be considered as an extraordinary and compelling reason for purposes of § 3582(c)(1)(A). Congress could not have been clearer: no matter how many GED certificates an inmate earns or how many drug treatment programs they complete, rehabilitation alone does not justify early release. This provision reflects a deliberate policy choice—the recognition that rehabilitation is the norm, not the exception, and that if every rehabilitated inmate were released early, the entire concept of determinate sentencing would collapse. Yet in the years since 2018, a growing number of courts have effectively nullified this statutory command by treating rehabilitation as a "plus factor" that, when combined with something else, becomes extraordinary.

I have personally argued against this trend in multiple federal appellate briefs, and I can tell you that the logic is circular. Courts that grant compassionate release based on rehabilitation plus a minor health condition—say, well-controlled hypertension—are essentially saying that any inmate who behaves well in prison and has any medical issue at all qualifies for release. That is not what the statute says. The statute requires "extraordinary and compelling reasons," not "ordinary and compelling reasons." Hypertension affects millions of Americans, including the incarcerated population, and is routinely managed with medication. To treat it as extraordinary is to redefine the word beyond recognition. Moreover, this approach creates perverse incentives: inmates may feign medical conditions or exaggerate symptoms to qualify for release, while those with genuine but less fashionable conditions are left behind.

The broader danger here is that the expansion of compassionate release undermines the credibility of the entire sentencing process. When a judge sentences a defendant to 120 months, both the defendant and the public understand that the sentence means 120 months—subject only to good-time credits under 18 U.S.C. § 3624(b) and the narrowest of exceptions. If that same defendant can return to court after serving three years and argue that their rehabilitation plus a routine health issue entitles them to release, then the original sentence was never real. I have seen defendants openly strategize about which medical conditions to document in their prison medical records, knowing that a diagnosis of asthma or obesity might later be leveraged into a compassionate release motion. This is not justice; it is a loophole that Congress never intended. The Sentencing Commission and the courts must return to the statutory text and enforce the plain meaning of "extraordinary and compelling."

Why the BOP's Administrative Exhaustion Requirement Has Become a Hollow Formality

The First Step Act's exhaustion requirement at 18 U.S.C. § 3582(c)(1)(A) was designed as a gatekeeping mechanism: an inmate must either exhaust all administrative remedies with the Bureau of Prisons or wait 30 days after the warden's receipt of a request before filing a motion in court. This requirement was intended to give the BOP an opportunity to evaluate compassionate release requests on the merits and to maintain institutional expertise in determining what circumstances truly warrant early release. In practice, however, the exhaustion requirement has become a meaningless procedural hurdle that inmates can circumvent with ease. I have seen cases where inmates file a request with the warden on a Friday, then file a motion in federal court the following Monday, arguing that the 30-day period has not yet expired but that the BOP's "constructive denial" justifies immediate judicial intervention. Too many courts accept this argument, effectively reading the exhaustion requirement out of the statute.

The Second Circuit's decision in United States v. Saladino, 7 F.4th 120 (2d Cir. 2021), while not a binding precedent on other circuits, opened the door to this approach by holding that courts may waive the exhaustion requirement when it would be futile or when the BOP has already indicated it will deny the request. Since then, a flood of motions has bypassed administrative review entirely. This is a dangerous departure from the statutory scheme. The BOP has institutional knowledge about prison conditions, medical resources, and public safety that individual district judges lack. When courts short-circuit the exhaustion process, they make decisions without the benefit of that expertise, often based on incomplete or self-serving information from the inmate. I have reviewed cases where a judge granted compassionate release based on an inmate's claim of inadequate medical care, only to discover later that the BOP had already scheduled the inmate for specialized treatment—a fact the court never learned because the exhaustion process was skipped.

The erosion of the exhaustion requirement also undermines the BOP's ability to manage its population effectively. Prison administrators rely on knowing which inmates are seeking release and for what reasons. When inmates can bypass the BOP and go directly to court, the agency loses its ability to plan for bed space, staffing, and medical resource allocation. Furthermore, the exhaustion requirement serves an important screening function: it filters out frivolous requests before they reach the federal judiciary. In my experience, the BOP denies the vast majority of compassionate release requests because they do not meet the statutory standard. Those denials are often correct. But when courts allow inmates to skip the administrative process, they are forced to adjudicate thousands of meritless motions that the BOP would have resolved quickly. This is a waste of judicial resources and a disservice to inmates with genuinely compelling claims who are stuck in a system clogged with frivolous filings.

If you or a loved one are considering a compassionate release motion, or if you are facing federal sentencing and want to understand how recent changes to § 3582(c)(1)(A) might affect your case, do not rely on general advice or internet research. The law in this area is shifting rapidly, and what works in one district may be reversible error in another. My firm has decades of combined experience in federal criminal defense, both at the trial and appellate levels, and we have successfully navigated the compassionate release landscape for clients across multiple circuits. Contact our office today for a confidential consultation—we will evaluate your circumstances, assess the strength of any potential motion under the current legal framework, and provide you with an honest assessment of your options. Sentencing certainty matters, but so does zealous advocacy within the bounds of the law. Let us put our experience to work for you.