Blog
Timing Your Cooperation: Before or After Indictment?
Cooperation is the only currency a federal defendant controls, and its value depreciates by the hour. The question is not whether to cooperate. For most defendants confronting serious federal exposure, the arithmetic on that question resolved itself the moment the investigation became visible. The question is when.
And the answer depends on information the defendant does not yet possess, which is the central difficulty of every cooperation decision and the reason most defendants arrive at it poorly. The government knows what it has. The defendant, at the stage when cooperation would be most valuable, knows almost nothing about the government’s case. That asymmetry is where the real decision begins.
The Proffer as Threshold
Before cooperation assumes any formal shape, it passes through a proffer session. The defendant sits with counsel in the United States Attorney’s office and provides what the government describes as a complete and truthful account of criminal activity: the defendant’s own conduct, the conduct of others, and anything adjacent to the investigation the government considers relevant. The agreement governing this session offers a species of protection that is narrower than most defendants perceive it to be.
The government agrees not to use the defendant’s statements in its case in chief. It does not agree to disregard those statements. Every lead the statements generate, every document identified, every witness located through information the defendant provided can be developed and introduced without restriction. The proffer functions, in this respect, the way a signed confession functions in a room where no one expected one: the record absorbs it, and the record does not forget. The defendant who proffers without grasping this distinction has already committed the first consequential error of the case.
Pre-Indictment Cooperation and Its Costs
The conventional advice extends as far as it is able: cooperate early. The Supreme Court in Wade v. United States confirmed that the government holds a power, not a duty, to file a substantial assistance motion under §5K1.1 of the sentencing guidelines. Among the five factors courts weigh is the timeliness of the defendant’s assistance.
A defendant who approaches the government before indictment occupies the strongest negotiating position the federal system permits. The government’s case may still be under construction. The defendant’s information can shape the direction of the investigation, identify new targets, produce evidence the government could not otherwise obtain, and in certain cases result in no charges being filed at all.
The risk is less often examined. A defendant who cooperates before indictment cooperates before discovery. The government bears no obligation to disclose the strength of its evidence at the proffer stage, and it does not. The defendant is, if we are being precise, committing to the most consequential decision of the case while possessing the least information about it. The earlier one cooperates, the greater the potential value to the government and the greater the potential departure from the guidelines. The earlier one cooperates, the less one knows about what the government already has, what it still needs, and whether the defendant’s information is something the government has already obtained from another source.
Three cases in the past two years arrived at this office after early proffers that revealed more than the government had assembled on its own. In each instance, the defendant disclosed conduct the government had not previously identified. The guidelines range was then calculated using that conduct. The cooperation produced a departure, but the departure began from a higher floor than the original exposure would have set.
The proffer agreement’s protection against direct use is real. The protection against derivative use is, in most districts and under most agreements, a sentence the defendant reads once and does not fully comprehend until the trial.
Our approach to this problem is to conduct, before any proffer session, what amounts to a reverse assessment: an effort to determine what the government is likely to know based on the investigative steps that have become visible. The subpoenas served, the search warrants executed, the witnesses contacted, the co-defendants whose attorneys have stopped returning calls. In some districts, the way the clerk’s office handles sealed filings or the way a particular AUSA conducts an initial meeting can disclose more than the government intended. That assessment does not yield certainty. It produces a range of possibilities, and the decision whether to proffer is made against that range rather than against the hope that the government knows less than it does.
The difference between cooperating in the first sixty days and cooperating at month four is often the difference between avoiding indictment and receiving a departure that brings a guidelines sentence from one figure to a slightly lower one. But the difference between cooperating with preparation and cooperating without it can be, in certain cases, the distance between a sentence that rewards the cooperation and a sentence that punishes the candor. The preparation is what separates a proffer that produces a departure from one that produces a confession.
After the Indictment
Once the indictment is returned, the defendant’s position changes in a way that the statute does not fully acknowledge. The government has committed to prosecution. The grand jury has found probable cause.
The defendant’s information, whatever its quality, now arrives in a context where the government has already assembled enough to proceed without it. The cooperation is no longer shaping the investigation. It is supplementing a case the government considers sufficient. §5K1.1 does not distinguish between pre-indictment and post-indictment assistance in its language. The timeliness factor, though, constructs a hierarchy in practice. By six months after indictment, cooperation begins to resemble a concession.
Rule 35(b) of the Federal Rules of Criminal Procedure permits the government to move for a sentence reduction based on substantial assistance provided after sentencing. The window is one year, with narrow exceptions for information the defendant could not have known at the time of sentencing. Whether the court intended this provision as a mechanism for late cooperators or as a procedural tool for ongoing investigations is a question worth considering.
The defendant who waits is not always making a calculated choice. Sometimes the delay is the product of something less strategic: the weight that accumulates during a federal prosecution, the conversations with counsel that begin to circle the same ground, the slow recognition that the case will not resolve on its own. The discovery materials arrive in boxes, and the boxes keep arriving.
Assessing Information Value
The most frequent miscalculation concerns not timing but value. Defendants believe they possess more than they do. The government does not need additional witnesses to facts it has already established through recordings and documents. It needs information that reaches targets it cannot access without the cooperator’s knowledge, and it needs that information before the case has solidified.
If the defendant’s information moves laterally (confirming what other sources have already provided) rather than vertically (reaching individuals the government cannot otherwise reach), the cooperation will produce a proffer session, an acknowledgment of guilt, and a reduction that falls short of what the defendant expected. There are exceptions, though in practice they tend to confirm the rule. This is the system performing its intended function.
Prosecutorial Discretion and §5K1.1
The architecture of federal cooperation places a single party in control of the outcome, and it is not the defendant. Under §5K1.1, only the government may file a substantial assistance motion. Under 18 U.S.C. §3553(e), only the government may move the court below a mandatory minimum. The judge cannot initiate that departure.
Wade v. United States established that courts retain the authority to review a prosecutor’s refusal to file a substantial assistance motion, but only where the defendant makes a showing that the refusal rested on unconstitutional grounds: the defendant’s race, religion, or a comparable factor. The practical threshold for that showing is, by the Court’s own language, prohibitive. A claim that the defendant merely provided substantial assistance does not entitle the defendant to a remedy or to an evidentiary hearing.
The cooperation decision is therefore a wager placed on the government’s future assessment. The defendant provides truthful information, subjects himself to the risks that accompany being identified as a cooperator, waives trial rights that cannot be recovered after the proffer, and then waits for the government to determine whether the assistance was substantial. The word appears nowhere in the guidelines with a fixed definition.
What we have observed (and this is an observation drawn from practice rather than from a dataset one would call rigorous) is that the government’s assessment of value correlates as much with the AUSA’s caseload and the district’s enforcement priorities as with the quality of the defendant’s information. The AUSA (who, in most cases, has formed this assessment before receiving so much as a telephone call from defense counsel, and whose calendar and institutional priorities will determine whether the cooperation is treated as an asset or an inconvenience) controls the only mechanism through which the defendant’s assistance can be recognized at sentencing. A cooperator providing the same information in the Eastern District of New York and in a smaller district with fewer active investigations may receive different treatment. The guidelines contemplate uniformity, but the districts do not always provide it.
Cooperation in the federal system is not a negotiation between equals. It is a surrender of information in exchange for a possibility, governed by a timeline that punishes hesitation and a framework that concentrates discretion in the hands of the party who needs the defendant least. The decision about when to cooperate is a decision about what one is willing to disclose before knowing what one is disclosing it into.
A consultation is where this question receives the specificity it requires. There is no cost to the first conversation, and no assumptions embedded in it. What there is, in most cases, is a set of facts that has not yet been organized and a timeline that has already started to move.