Blog
After Cooperation
The cooperation is finished, and nothing has changed. The proffer sessions are complete. The debriefings have been transcribed, the testimony delivered, the names provided. The government has what it required. The cooperator has a plea agreement, a sentencing date that has passed, and a sentence that may or may not reflect the assistance rendered. What remains is silence, and silence in the federal system is not reassurance.
The government will not call to confirm that the information was useful, and no timeline exists for what follows. There is a particular stillness to this phase of the process, familiar to anyone who has sat in it: the case file closed on the government’s end, the cooperator’s life suspended on a question only the prosecutor can answer.
You signed the agreement because someone explained that it was the only option remaining that did not involve a statutory minimum measured in decades. You provided what was asked. The question you are now living inside, the one that occupies every conversation with counsel and every calculation about the future, is whether what you provided will be recognized as enough. The government has no obligation to tell you when, or whether, that call will come.
The Government’s Sole Authority to File
Under Section 5K1.1 of the Sentencing Guidelines, only the government may move the court for a downward departure based on substantial assistance. Defense counsel cannot file the motion. The court cannot initiate it. The cooperator, regardless of the quality or quantity of assistance provided, possesses no procedural mechanism to compel the government to acknowledge what was given.
Defense counsel can structure the cooperation, prepare the proffer, document every session, and argue at sentencing that the assistance was significant. What counsel cannot do is substitute for the government’s motion. The architecture of the system places the prosecutor in a position of, if we are being precise, unilateral authority over the cooperator’s sentencing outcome.
The court held in Wade v. United States that a refusal to file could be reviewed only if grounded in an unconstitutional motive, which in practice constitutes a standard almost no defendant satisfies. The unconstitutional motive threshold requires the defendant to demonstrate that the government’s refusal was based on race, religion, or another prohibited classification. A claim that the government acted arbitrarily, or that it undervalued the cooperation, or that it simply changed its assessment of the information’s usefulness, does not reach that threshold.
We have seen prosecutors decline to file after a cooperator testified at trial, took considerable personal risk, and provided information that resulted in a conviction. We are aware of three cases in the past eighteen months where the government’s stated reason for declining to file was that the information “duplicated” what another cooperator had already provided. In one of those cases (which involved a cooperator who had worn a recording device on seven occasions and whose recordings were played at trial) the duplication rationale was difficult to credit, though the government’s discretion on the point was functionally absolute.
The agreement says “may.” That word is the entire architecture of the relationship.
What Counts as Substantial
The word “substantial” performs more labor in federal criminal practice than any other adjective in the Guidelines. The statute does not define it. The Guidelines commentary identifies factors the court should consider, including the significance, usefulness, truthfulness, and timeliness of the assistance. None of these factors are weighted. None of them are dispositive.
A cooperator who provides the name of a supplier, wears a recording device for three controlled purchases, and testifies before a grand jury may receive a 5K1.1 motion or may not, and the factors that distinguish the two outcomes are not always articulable. The determination belongs to the prosecutor and, though the court may review it for unconstitutional motive, the practical threshold for that review remains so elevated that the discretion functions as unreviewable.
What permitted one cooperator to receive a departure of something like forty percent below the guideline range while another, in the same district, with a comparable scope of assistance, received nothing is a question that the system is not designed to answer transparently. The Sentencing Commission publishes aggregate departure data. It does not publish the reasoning.
Whether the same testimony would have been valued in a different district, with a different AUSA, is a question we cannot resolve from the outside.
The Silence After Sentencing
In the months after sentencing, the cooperator enters a period the statute does not name. The formal cooperation may be complete. The government’s assessment of its value may not be. For cooperators whose assistance continues after sentencing, Rule 35(b) of the Federal Rules of Criminal Procedure provides a mechanism for sentence reduction, but the motion belongs, again, to the government.
Rule 35(b) permits the government to move for a sentence reduction based on post-sentencing substantial assistance, but the motion must generally be filed within one year, and the exceptions to that deadline, while they exist, are construed with a narrowness that leaves little room for the cooperator who provided information the government was slow to use, or whose assistance bore fruit in an investigation that did not produce an indictment until fourteen months after the original sentencing date. There are exceptions, though in practice they tend to confirm the deadline rather than relieve it.
The cooperator, meanwhile, is serving the original sentence. The reduction that was discussed, that was the reason the cooperation was undertaken, that sustained the cooperator through the risk of discovery by codefendants and the particular stress of wearing a recording device into a room where violence was a possibility: that reduction has not materialized. It exists as a contingency in a plea agreement, controlled by a party with no obligation to act on it.
Most people who cooperate do not understand what the waiting will feel like until they are inside it.
I drafted portions of this piece on a Friday afternoon in late winter, after a call with a client who had been waiting nine months for a Rule 35 filing the government had verbally indicated was forthcoming. The filing had not arrived. The client’s release date, calculated without the reduction, was still four years away.
That uncertainty is not a flaw in the system.
The government considers it a feature, because continuing authority over the cooperator ensures continued availability for testimony, debriefings, and whatever else the investigation requires. A cooperator who has already received the sentencing benefit has less incentive to remain available. A cooperator who is still waiting has every incentive in the world. Whether the government consciously calibrates the timing of its Rule 35 filings to maintain this hold is a question worth considering.
The phone call from the AUSA’s office, when it comes, does not always contain what the cooperator expected. The recommended departure may be smaller than discussed. The motion may acknowledge the assistance but characterize its significance in terms the cooperator finds unrecognizable. In some districts, the government’s recommendation to the court takes the form of a letter that the cooperator and defense counsel do not see before it is filed.
Six months is not unusual.
Safety, Disclosure, and the Record That Follows
Cooperation, once undertaken, becomes a permanent feature of the legal record. The cooperation agreement itself constitutes Giglio material: information bearing on a witness’s credibility that the prosecution is constitutionally required to disclose in any subsequent proceeding. If the cooperator is ever called to testify again, in any context, the fact of the prior agreement and its terms will be available to opposing counsel.
This disclosure obligation extends beyond criminal proceedings. The cooperator’s status can surface in civil litigation, regulatory matters, and employment disputes in ways that are not always foreseeable at the time the agreement is executed. The credibility question, once opened by the cooperation, does not close when the case that prompted it is resolved.
For cooperators who are incarcerated, the Bureau of Prisons presents its own difficulties, and the Task Force on Protecting Cooperators has acknowledged the problem of violence directed at those perceived to have assisted the government. The Task Force recommended changes to electronic filing procedures and to BOP protocols, some of which have been adopted and some of which remain under consideration. The court system has moved toward restricting remote access to plea and sentencing documents, which may limit the ability of other inmates to identify cooperators through PACER searches. Whether these measures are sufficient is an empirical question the system has not yet answered.
18 U.S.C. Section 1513 imposes severe penalties for retaliation against witnesses and informants, and the statute carries extraterritorial jurisdiction. The penalties for retaliatory violence extend to life imprisonment. The statute exists, and it is enforced, but its existence does not prevent the underlying conduct from occurring.
A cooperator who later seeks employment in any regulated industry, or who becomes involved in subsequent litigation, carries the fact of the cooperation in ways that are not always visible at the time the agreement is signed. That history remains accessible for as long as the legal system retains interest in the cooperator’s credibility.
What This Firm Does at the Outset
Before a client enters a proffer session, we construct the cooperation as if the government will not honor it. This is not pessimism. It is the only responsible posture given the discretion the government retains.
The plea agreement in our cases specifies, where the AUSA will accept it, that the government “will” file a 5K1.1 motion upon satisfactory completion of cooperation, rather than that it “may” consider doing so. The difference between those two words is the difference between a contractual obligation and an aspiration. Not every prosecutor will agree to the stronger language. We ask for it in every case, and the frequency with which it is refused tells you something about how the government perceives the relationship.
We document every debriefing, maintain a contemporaneous record of information provided, and ensure that the client’s contribution is memorialized in a form the court can review independently of the government’s characterization. If the government declines to file, and the plea agreement contains mandatory filing language, we are prepared to bring a breach of plea agreement claim. If the agreement contains only discretionary language, we preserve the record so that the court may still consider the cooperation under the Section 3553(a) factors at sentencing, as the Second Circuit recognized in United States v. Fernandez.
I am less certain than most defense attorneys that early cooperation is always preferable to waiting; the timing depends on the government’s investigative posture and on how many other cooperators are already in the pipeline. In some cases, the cooperator who arrives second provides less value not because the information is less significant but because the government has already obtained it from someone who arrived first. The sequence matters. We assess it before the first proffer.
This is not standard practice, and it requires more from the client at the front end of the process. The preparation is longer. The documentation is heavier. The conversations about risk are more detailed than what most firms provide, because we have observed what happens when a cooperator enters the process without a record sufficient to hold the government accountable. A consultation is where that conversation begins, and it assumes nothing beyond the hour.