Blog
Can My Cooperation Be Used Against Me in State Court?
The Short Answer Is Yes, and the Longer Answer Is Worse
A federal proffer agreement does not bind a state prosecutor. One may cooperate with the United States Attorney’s Office under a written agreement that restricts the federal government’s use of the statements provided, and still discover that a state district attorney has obtained those statements, or evidence derived from them, through entirely lawful channels. The constitutional protections that govern compelled testimony under formal immunity do not extend, in the same form, to the voluntary disclosures made under a proffer letter.
This is, if we are being precise, not a gap in the law so much as a feature of it. The federal system and each state system operate as separate sovereigns. A contract between a defendant and one sovereign creates no obligation on the part of another. Most clients who sit down for a proffer session understand this in theory. They do not understand it in practice until the state files charges.
The instinct to cooperate is rational. Federal sentencing guidelines reward substantial assistance, and a 5K1.1 motion from the government can reduce a sentence below the mandatory minimum. The calculus changes, however, when the cooperation itself generates material that a separate sovereign can use without constraint. The agreement you signed was between two parties, and the state was not one of them.
The Architecture of a Proffer Agreement
The proffer letter, sometimes called the Queen for a Day agreement, is a contract. It establishes that the defendant will provide truthful information to federal prosecutors and agents. In return, the government agrees not to use the defendant’s statements as direct evidence in its case. The meeting takes place at the United States Attorney’s Office, with counsel present, and the session is documented.
The protections, such as they are, divide into two categories. The first is use immunity: the government will not introduce the defendant’s own words against the defendant at trial. The second is the Kastigar waiver, which most proffer letters now require. Under Kastigar v. United States, a grant of formal immunity would require the government to prove that all evidence in a subsequent prosecution was derived from sources wholly independent of the immunized testimony. The Kastigar waiver eliminates this burden. The defendant, by signing, agrees that the government may make derivative use of the information provided. They do not sign it because they understand it. They sign it because they have been told that cooperation is the path to leniency. That is what makes it dangerous.
The derivative use provision functions the way a smoke detector functions in a building that has already been condemned: technically present, operationally meaningless. The government may follow any investigative lead the proffer generates. It may locate witnesses, subpoena records, and develop evidence that flows from the defendant’s own disclosures. The language is standard, the format familiar, the consequences rarely discussed in sufficient detail.
If the defendant later testifies in a manner inconsistent with the proffer, the statements may be used for impeachment. If the defendant is found to have been untruthful, the agreement dissolves. If no formal cooperation agreement follows, the government retains whatever investigative advantage the proffer session provided. The derivative use provision is the one that causes damage, and it is the one most clients do not read until it is too late.
Statutory Immunity and Its Cross-Sovereign Reach
Formal statutory immunity operates under different constitutional constraints than a proffer letter. Under 18 U.S.C. § 6002, a federal court may compel testimony by granting use and derivative use immunity. The compelled testimony, and any evidence derived from it, may not be used against the witness in any criminal case. The Supreme Court in Kastigar held that this standard is coextensive with the Fifth Amendment privilege.
The cross-sovereign question was addressed earlier, in Murphy v. Waterfront Commission. The Court held that incriminating testimony compelled by one government may not be used by another. A state witness immunized under state law could not have that testimony used by federal prosecutors, and the reverse held as well. The rule, in practical terms, is that compelled testimony in one jurisdiction cannot be weaponized by another.
This principle applies to formal, statutory grants of immunity. The Murphy framework applies to formal, compelled testimony under statutory grants. It was constructed around the Fifth Amendment’s prohibition on compulsory self-incrimination, and its logic depends on the testimony having been compelled in the first instance.
A proffer agreement is not compelled testimony. It is a voluntary statement made under a contractual arrangement with a single prosecuting office. The constitutional architecture that prevents cross-sovereign use of immunized testimony was not designed for, and has not been extended to, voluntary disclosures made under a Queen for a Day letter. The distinction is not academic. It determines whether a state prosecutor is constitutionally barred from using the information, or merely uninvited from the federal agreement.
The case law on this question is less settled than one would expect for a problem this common. Some courts have treated the proffer letter’s protections as binding only on the signatory (the federal government), leaving the state entirely free to use the statements or their fruits. Other courts have examined the degree of coordination between state and federal prosecutors and, where the state functioned as an arm of the federal investigation, have applied broader protections.
Whether the court intended this outcome or merely failed to prevent it is a question worth considering.
The Gap Between Federal Promises and State Prosecution
In most of the cases that reach this office, the client believed the proffer agreement covered everything. They believed the word “government” in the letter meant all government, every level, every jurisdiction. It did not.
What they signed was a federal contract, not a constitutional shield. The federal prosecutor agreed not to use the defendant’s statements in the federal case. The agreement said nothing about the state, because the federal prosecutor has no authority to bind the state. A state prosecutor who obtains the proffer material, whether through a parallel investigation, a shared database, or a referral from a federal agent, is free to use it. The restriction that applied in the federal case does not follow the information into state court.
A state prosecutor, reviewing a file that federal agents assembled in part from the defendant’s own admissions, is under no contractual or constitutional obligation to disregard what she finds there. Something like forty percent of the time, the state file references material the federal proffer generated. This figure is not scientific; it comes from what we have observed in cases involving parallel investigations, particularly in financial fraud and narcotics matters, where federal and state interests overlap. The overlap is the problem. When a federal proffer identifies witnesses, accounts, or patterns, the resulting evidence enters the record. A state investigator working the same case can access that evidence through standard channels.
I am less certain about this than the preceding analysis might suggest, because the case law in this space remains thin and the circuits do not all agree on where Murphy ends and contract begins. In the Second Circuit, some decisions have examined the degree of federal and state entanglement before allowing the state to use proffer-derived evidence. In other jurisdictions, the question has not been addressed with any precision. The absence of clear precedent is itself a risk.
The agreement sat in a drawer for six months while the state assembled its own case. Three cases this year alone involved state indictments that cited evidence traceable to the federal cooperation. In each, the client had been assured by prior counsel that the proffer agreement would prevent this outcome. It did not prevent anything outside its own four corners. The state had its own file, its own timeline, its own reasons.
Dual Sovereignty After Gamble
The Supreme Court in Gamble v. United States confirmed what practitioners had long assumed. The federal government and each state derive their prosecutorial authority from independent sources, and an offense against one is not an offense against the other. A person may be prosecuted by both sovereigns for the same conduct without violating the Double Jeopardy Clause. The decision was seven to two, and it closed the door on what had been a recurring constitutional challenge.
For cooperators, Gamble means that a federal plea and a state prosecution for the same underlying conduct can coexist. The Petite policy (under which the DOJ exercises internal restraint against successive federal prosecutions where state charges have already been resolved) is an administrative guideline, not a constitutional requirement, and it does not restrict state prosecutors at all.
In practical terms, a defendant who cooperates at the federal level and pleads to federal charges retains full exposure at the state level. The state may prosecute the same conduct under its own statutes, and the federal cooperation, far from insulating the defendant, may have generated the very evidence the state requires. Gamble is not the obstacle; it is the terrain.
The Debt That Silence Cannot Discharge
By the time a state prosecutor receives a cooperator’s statements, the damage is structural. By then, the information is not a statement; it is evidence. The proffer identified witnesses. The witnesses gave their own statements. The statements led to records. The records corroborated the conduct. The cooperator’s original words are no longer necessary for the state to proceed.
The letter from the state AG’s office arrives on its own schedule, usually a Tuesday or a Thursday, and it references conduct the client assumed had been resolved months earlier. The client calls the federal attorney, who explains that the proffer agreement applies to the federal case. The client calls this office.
The timing matters. A motion to suppress the state’s use of proffer-derived evidence must be filed within the state court’s procedural deadlines, and the argument itself depends on demonstrating a factual chain between the proffer and the evidence in the state’s file. That chain is not always visible from the defense side without discovery. The filing deadline does not adjust for regret.
What We Do Differently in Cross-Sovereign Matters
Before any client in this firm enters a proffer session, we address the state exposure in writing. We send a letter to the relevant state prosecutor’s office identifying the cooperation, requesting confirmation that the state will not use proffer-derived material in any pending or future state proceeding, and documenting the date and scope of the federal agreement. We write that language ourselves, and we do not proceed until the response is in hand.
This is not standard practice. The standard approach is to sign the federal proffer letter as presented, conduct the session, and address state-level consequences only if they arise. We find that approach inadequate, not because it is wrong in theory, but because we have seen what happens when the state files arrive. The result is a cooperation framework that accounts for the existence of more than one sovereign, which is something the standard federal proffer letter does not do.
We also review the proffer letter’s derivative use clause with a specificity that the form letter does not invite. In cases where state charges are pending or probable, we negotiate language that restricts the federal government’s ability to share proffer material with state agencies. Not every United States Attorney’s Office will agree to this language. Some will. The ones who refuse are communicating something about the investigation that the client is entitled to know before deciding whether to cooperate.
The standard approach treats cooperation as a single-track problem; we regard it as a problem with at least two tracks, and sometimes three. A first consultation costs nothing and assumes nothing; it is where the diagnosis begins.