Blog
What Is a Federal Proffer Agreement ("Queen for a Day")?
The proffer agreement is the most misunderstood document in federal criminal practice. Defendants sign it believing their words cannot follow them into a courtroom. Prosecutors draft it knowing how those words will travel. Between the text of the letter and the reality of what follows, the distance is considerable, and most of the danger lives in that distance.
The nickname, “Queen for a Day,” implies temporary sovereignty. For the duration of the session, one speaks freely, and what one says is shielded from use at trial. The phrase is older than most practitioners realize, borrowed from a television program that aired in the 1950s, though the connection to its current legal meaning is loose at best. What the name captures, and what it conceals in equal measure, is the conditional nature of the protection. The protection it confers is narrower than the phrase suggests, and the conditions attached to it do most of the work.
Federal Rule of Evidence 410 and Federal Rule of Criminal Procedure 11(e)(6) provide the statutory foundation. Statements made during plea discussions with a prosecuting attorney are, under those provisions, inadmissible against the defendant if the discussions do not produce a guilty plea. That is the baseline. The proffer letter rewrites it.
The Terms of the Letter
A proffer agreement is a written contract between a federal prosecutor and an individual under criminal investigation. It permits the individual to disclose information about criminal activity in exchange for a circumscribed assurance: the government will not use that individual’s statements in its case-in-chief at trial. The assurance is genuine. It is also the narrowest form of protection the federal system offers.
The session takes place, in most cases, at the local United States Attorney’s Office. Defense counsel, the AUSA (who, in cases involving parallel civil proceedings, may share the room with SEC enforcement counsel or inspectors general from one or more federal agencies, a development that has altered the dynamics of the proffer session in ways that the original Rule 410 framework did not anticipate), and one or more federal agents are present. The defendant answers questions. Sessions can last four or five hours, and in complex cases they extend across multiple days.
The letter itself is two or three pages. It recites that the session does not constitute a plea discussion, and that the defendant voluntarily waives any right under Rule 410 or Rule 11(f) to suppress statements made during the interview. That waiver became standard after the Supreme Court’s 1995 decision in United States v. Mezzanatto, which held that a defendant may waive Rule 410 protections provided the waiver is knowing and voluntary. Prosecutors in most districts will not proceed with a proffer session without it.
What survives the waiver is the government’s commitment not to introduce the defendant’s own words during its direct presentation of evidence. This is a real constraint, but it governs the case-in-chief and almost nothing else. Impeachment, rebuttal, sentencing proceedings, derivative investigation: the letter permits all of these. The exceptions consume most of the protection the agreement appears to provide.
A defendant considering a proffer confronts a document whose protections are genuine on their face and limited in their operation. The holes are not concealed. They are written into the agreement in plain terms. The problem is that plain terms, in a letter this consequential, require a reader who understands what each exception permits the government to accomplish in practice.
Derivative Use and the Kastigar Waiver
The space between “we will not use your words” and “we will not use what your words teach us” is where most of the danger resides.
Every standard proffer agreement reserves the government’s right to make derivative use of information disclosed during the session. The language varies by district, but the substance is uniform. The government may pursue any investigative leads suggested by the defendant’s statements, and any evidence obtained from those leads may be used against the defendant in any future proceeding. The principle traces to Kastigar v. United States, the 1972 Supreme Court decision that distinguished use immunity from transactional immunity. Under Kastigar, use immunity prevents the government from introducing a witness’s compelled testimony at trial but does not prevent the government from following the threads that testimony reveals. The proffer agreement applies the same logic to voluntary cooperation.
In practice, this means a defendant who mentions a meeting, a bank account, a co-conspirator’s name, or a particular document during a proffer has not merely disclosed information. The defendant has provided a map. The government will subpoena the bank records. It will interview the co-conspirator. It will obtain the document through channels that can be characterized as independent of the proffer, and the resulting evidence is admissible without restriction. Whether the court intended this architecture or merely failed to prevent it is a question worth considering.
The derivative use provision is, if we are being precise, not a loophole at all. It is the design. Prosecutors do not enter proffer sessions expecting to use the defendant’s statements at trial. They enter expecting to learn where the evidence is. The session functions as an interview, and the immunity is the cost of the interview, and what the government purchases with that cost is direction. A name, a date, a location the agents had not yet identified.
There is a particular silence in a conference room at four in the afternoon, after the AUSA has asked the same question three different ways, and the defendant has answered it each time with slightly different details. That silence is diagnostic. The agents are not confused. They are cataloguing inconsistencies. If the defendant later testifies at trial in a manner that departs from the proffer, those statements become impeachment material. If the departure is substantial enough, the government will argue, with the letter’s own language in support, that the defendant’s proffer statements should be received as substantive evidence in rebuttal.
Mezzanatto opened this door. The Supreme Court held that a defendant who agrees to waive Rule 410 protections for impeachment purposes has entered a valid and enforceable contract. Justice Ginsburg, concurring, expressed concern that extending the waiver to the government’s case-in-chief could discourage plea discussions entirely. Several circuits have extended the holding in that direction. The Second Circuit, in United States v. Barrow, concluded that proffer agreement waiver provisions could permit use of statements in the government’s direct case. The D.C. Circuit reached a similar conclusion in United States v. Burch. The letter that promised protection has become, in these jurisdictions, a vehicle for the admission of the defendant’s own statements.
The defendant who proffers is choosing between cooperation and contestation. One cannot do both. The proffer locks the defendant into a version of events. If that version is truthful and complete, the lock serves the defendant’s interests, because the path forward is cooperation, a plea, and, in the best case, a government motion under Section 5K1.1 of the Sentencing Guidelines for a downward departure. If the version is incomplete, or if the defendant’s memory is imperfect on even one material point, the lock becomes a trap.
And the false statement risk is not hypothetical. Section 1001 of the federal criminal code makes it a federal offense to provide false statements to the government. A defendant who misremembers a date, confuses one meeting with another, or provides an answer that contradicts a document the government has not disclosed can face prosecution for the misstatement. In practice, standalone Section 1001 charges arising from proffer sessions are rare. But the threat shapes the session. The agents know the documentary record, and the defendant, more often than one would prefer, does not.
The Room Itself
The physical setting of a proffer session carries its own weight. Government offices are designed for this purpose, though not in the way an architect would intend. The security procedures, the windowless conference room, the recorder on the table: all of it communicates that the defendant is operating under the government’s terms. Some defense attorneys try to hold the session at their own offices. The difference in a client’s composure can be real, though difficult to measure.
The first letter from a prosecutor arrives without ceremony. It arrives on a Thursday or a Friday, which is not a coincidence. The letter outlines the terms under which the government will conduct a proffer and asks defense counsel to contact the AUSA to schedule the session. For many defendants, particularly those in white-collar investigations who have never encountered the criminal system, the letter is the first confirmation that the government’s interest is serious and specific.
What the client does between that letter and the session matters more than most clients understand. The wrong move is to agree to proffer out of panic, before counsel has reviewed the evidence and assessed the government’s position. The other wrong move is to refuse cooperation before understanding what the government already possesses. Both errors are common, and both are expensive.
When Cooperation Serves the Client
The proffer is a means, not an end. Its value depends on what follows.
If the defendant provides information that is truthful, complete, and useful to the government’s investigation of other individuals, the proffer can open a path to a cooperation agreement. That agreement, in turn, can produce a government motion under Section 5K1.1 of the Sentencing Guidelines, requesting that the court depart downward from the applicable guideline range. Only the government can file this motion. The defendant cannot compel it.
A 5K1.1 motion is the most consequential sentencing tool in the federal system. Reductions vary across districts, but the effect can be substantial, particularly for defendants facing mandatory minimums. In cases where the government also files a motion under 18 U.S.C. § 3553(e), the court may sentence below the statutory floor. Without both motions, the mandatory minimum holds, and the proffer’s value is limited to whatever goodwill the government extends at sentencing.
The decision to proffer is a decision about information. Does the defendant possess knowledge the government does not already have? Can that knowledge lead to the investigation or prosecution of individuals the government considers more culpable? If both answers are yes, cooperation has a rational basis. If either answer is no, the calculus changes. The defendant may be offering the government a roadmap to evidence that strengthens the case against the defendant, without receiving anything of value in return.
We evaluate proffer decisions by working backward from sentencing. The question is not whether the client can provide information. The question is whether the information the client possesses is likely to produce a cooperation agreement, whether that agreement will result in a 5K1.1 motion, and whether the resulting sentencing reduction outweighs the risks the proffer creates. The analysis is specific to the case, the district, the AUSA, and what the government has already assembled. There are exceptions to this framework, though in practice they tend to confirm the approach.
I am less certain about how proffers function in districts where the U.S. Attorney’s Office applies cooperation credit with less consistency; the Southern District of New York and the Eastern District of Virginia, for instance, have reputations for different levels of generosity with 5K1.1 motions, and the gap between those offices shapes the advice we provide to clients in multi-district investigations. What I can say, from something like twelve years of this work, is that a client who proffers without a clear cooperation target in mind is a client who has exchanged information for nothing.
The Decision Before the Session
Before any proffer session, the attorney’s obligation is to know as much as the government knows, or as close to that as discovery and pre-indictment discussions permit. The proffer is not the first conversation with the government. It is the last one before the client’s words belong, in some measure, to the prosecution.
The preparation involves several steps that must precede the session:
- Review all available evidence and identify the government’s probable theory.
- Work through the client’s recollection against the documentary record.
- Identify every point where memory diverges from documentation.
- Assess whether the client possesses information of genuine value to the government’s investigation of other targets.
Where memory diverges from documents, counsel and client resolve the gap before the session, not during it. An honest failure of memory, unaddressed, becomes a potential false statement. The distinction between “I do not recall” and “no” is the distinction between caution and exposure.
There is a broader principle at work in the proffer agreement, one that extends past any individual case. The federal system is built to reward cooperation and to penalize silence, and the proffer is the mechanism through which that design operates. Whether the mechanism places adequate limits on what the government may do with the information it receives, whether the protections correspond to what a defendant expects when signing the letter: these are open questions, and the case law has not settled them in the defendant’s favor.
A consultation is where this conversation begins. A first meeting with our office costs nothing, assumes nothing, and exists for a single purpose: to determine whether a proffer serves the client’s interests or whether silence serves them better. That determination requires a specific assessment, conducted under privilege, before any words are exchanged with the government.