§ 12.04. Documentation and Disclosure of Informant Evidence

(a) At a minimum, agencies should ensure documentation of:

  • (1) the informant’s name, age, demographic information, and prior criminal history;
  • (2) all current and previous arrangements with the informant, including any incentives offered;
  • (3) all statements made by the informant, including the substance thereof and the time and place they were made;
  • (4) information regarding any prior use of the informant or the informant’s testimony, including any prior cooperation agreements, benefits provided or offered, erroneous statements, recanted previous statements, and violation of agency guidelines for informants;
  • (5) whether the informant suffers from substance-abuse issues, mental-health issues, or other characteristics that would make the informant especially vulnerable or unreliable; and
  • (6) whether the informant was formerly or is currently involved in criminal activity.

(b) All conversations with informants should be recorded, when feasible, and otherwise should be documented as close to contemporaneously as possible.

(c) Agencies should share documentation of information concerning informants with prosecutors in an ongoing fashion, regardless of the status of an investigation or criminal case.


a. Documentation. Some of the work of informants is inherently difficult to document. Nonetheless, the safe and reliable use of informants requires careful documentation when feasible, given safety and practicality concerns. This makes it incumbent upon law enforcement to document carefully what it can, in writing or through video or audio recordings. Too often, the work that agencies have done with informants has been informal and poorly documented.

b. Informant information. As a first step, agencies should make a record of each person whom they rely upon as an informant, including name, age, demographic information, and prior criminal history. As part of documenting the identity of an informant, officers should ensure that they know the criminal history of that informant. All of this should occur as part of the prescreening described in § 12.02.

c. Informant agreements. All informant agreements should be reduced to writing. See § 12.01. Agencies often have entered into informal or oral agreements with informants, leading to a range of serious problems, including a lack of information about the scope of the agreements. Informants themselves may not adequately understand the agreement. Absent documentation, an informant can deny or distort the nature of an agreement.

Written informant agreements should reflect the complete agreement reached. Whenever a law-enforcement agency enters into an agreement with an informant, it should document all relevant terms of the agreement, including any incentives that the informant will receive, any prior use of the informant or prior testimony the informant has provided, and the informant’s identity. Such agreements should be documented contemporaneously so that there is a clear record of their terms and the existence thereof. Such agreements should be signed by both parties. Such documentation also will ensure that agencies can comply with their constitutional duties and will aid agencies in assessing the reliability of informant evidence. If the agreement with an informant changes, as it might for long-term informants, those changes too should be documented.

d. Informant statements. Consistent with the focus in these Principles on reliability, agencies always should seek to carefully document informant evidence. All statements by informants or conversations between law enforcement and informants should be documented and, when feasible, recorded. The need to document informant evidence is particularly great given reliability concerns. Documenting information that informants obtain in the field may pose practical and safety challenges, but when feasible, officers may use hidden recording devices or surveillance to produce more reliable information. When law-enforcement officers have conversations with informants at agency facilities or other secure locations, it should be routine to record those conversations. Although it may not be feasible to obtain full recordings in the cases of certain sensitive or more confidential informants—whose identities are protected even from some within law enforcement—it may be possible to make recordings by using only audio or by masking the voices or appearances of the informants. Similarly, in cases raising national-security concerns, recordings may pose heightened risks and may not be feasible. In such instances, however, careful and contemporaneous documentation, including concerning the safety reasons for concluding a recording is not feasible, can help ensure that agencies comply with their constitutional duties, and will aid them in assessing the reliability of informant evidence. Further, subsequent conversations may be feasibly recorded, in order to document the evidence and clarify any uncertainties in the prior, unrecorded conversations.

e. Prior use of informants. Agencies should document their prior use of informants. An informant may be a known quantity in an agency, and he or she may have given prior statements or provided information in prior cases, including formal courtroom testimony. Law-enforcement officers should be familiar with the performance of informants on past occasions. Evidence that an informant has recanted statements, or that information the informant provided in the past proved unreliable, is important to document. Such documentation can help ensure that agencies comply with their constitutional duties, as well as aid them in assessing the reliability of informant-provided evidence.

f. Vulnerable populations. Some informants may be members of vulnerable populations and therefore pose a special risk of danger to themselves. Members of vulnerable populations also may be at a special risk of being influenced more easily by suggestions made by law-enforcement officials, and therefore are more likely to provide unreliable information. Any information regarding an informant’s vulnerability, or any other information bearing on the reliability of an informant, should be documented.

g. Databases. Agencies should maintain databases to track information about informants, including their identities, their prior use by law-enforcement officials, any prior testimony they have provided, and other characteristics that can be tracked. Such a database can help an agency answer questions concerning the reliability of a given informant. Agencies similarly should track and document use of confidential informants, using (when necessary) identifying numbers or other methods to anonymize identifying information. For informants not formally labelled as confidential, but whose use raises additional safety or confidentiality concerns, agencies should create additional procedures to maintain the security of their identities and records.

h. Disclosure. Agencies have an obligation to carefully document any agreements they have with informants so that they can disclose such agreements to prosecutors. Prosecutors have disclosure obligations under criminal discovery rules. Prosecutors also have disclosure obligations in connection with probable-cause assessments in warrant affidavits, which are shared with the defense and with the court. For this reason, agencies must disclose to prosecutors any agreements with and benefits conferred on informants.

It is important to note that there should be ongoing disclosure to prosecutors of all information regarding arrangements with informants, including any incentives that they have or will receive, any prior testimony they have provided and the contents thereof, their use in prior investigations, and their identities. At the earliest time that it becomes relevant, this information should be available to prosecutors. Doing so enables prosecutors to carry out their ethical and constitutional duty to weigh whether they should disclose such information to the defense, as well as the probative value of evidence provided by informants. Such information may be provided on a redacted basis, due to concerns for informant safety, or submitted in camera for judicial review before disclosure to the defense.

Reporters’ notes

The uses of informants, ranging from informant identity, to agreements with informants, to the information provided by informants, traditionally have been documented in a highly informal manner. “Historically, many communications that occurred during the[] proffer sessions were not memorialized by government agents, especially preliminary sessions when prosecutors anticipated that the informant would be less than fully forthcoming.” Jessica A. Roth, Informant Witnesses and the Risk of Wrongful Convictions, 53 Am. Crim. L. Rev. 737, 753-754 (2016). There continues to be wide variation among the practices used by law enforcement and prosecutors’ offices to reduce arrangements made with informants to written agreements.

Far from promoting sound documentation of informant-provided and informant-related evidence, constitutional rules requiring the disclosure of written and formal arrangements may have had the opposite effect of encouraging informal and oral arrangements. R. Michael Cassidy, ‘‘Soft Words of Hope:” Giglio, Accomplice Witnesses, and the Problem of Implied Inducements, 98 Nw. U. L. Rev. 1129, 1132 (2004) (stating that “[t]he Court’s decision in Giglio[v. United States, 405 U.S. 150 (1972)] has created an incentive for prosecutors to make representations to an accomplice witness that are vague and open-ended, so that they will not be considered a firm ‘promise’ mandating disclosure. . . . Such indefinite agreements have the added advantage of allowing prosecutors to argue to the jury that no specific promise has been made to the witness”).

The lack of documentation of informant-related and informant-provided evidence has led to a wide range of problems. These include poorly understood, informal, and concealed arrangements with informants, as well as poorly recorded and uncorroborated statements by informants. In cases of persons later exonerated by post-conviction DNA testing, informants had sometimes falsely denied at criminal trials that they had reached arrangements with the prosecution. Brandon L. Garrett, Convicting the Innocent: Where Criminal Prosecutions Go Wrong 129 (2011). In various instances, an informant has claimed to have overheard details about a crime that only the culprit could have known; but without a recording of the informant’s interview(s) with law enforcement, it could not be determined whether those details had been disclosed to the informant by law enforcement. Id. at 131. There is evidence that informants delivered statements carefully crafted to incorporate the prosecution’s evidence and theories. Id. at 134 (“The most enterprising jailhouse informants did not just know specific facts about the crime. They knew the facts that the prosecutors had been unable to prove any other way. Their statements were neatly tailored to fit the prosecution strategy at trial.”). Such cases highlight the importance of recording, when possible, informant interviews and statements.

Agreements regarding the use of informants should be reduced to writing at the time the agreement is reached. Prosecutors often reach agreements with informants, but not necessarily with the involvement of law enforcement, and without making a written record. In recent years, “some offices (notably the U.S. Department of Justice) have changed that practice and now strongly recommend to prosecutors that notes be taken during all proffer sessions so that, among other reasons, they will be available to the defense at any trial at which the cooperator testifies.” Jessica A. Roth, Informant Witnesses and the Risk of Wrongful Convictions, 53 Am. Crim. L. Rev. 737, 754 (2016). To this day, “[m]ost prosecutors’ offices and law enforcement agencies, however, do not electronically record proffer sessions with potential informants.” Id. at 754. There is a need for coordination between law enforcement and prosecutors regarding the use of informants, and the documentation of agreements reached with informants. Both policing agencies and prosecutors’ offices should have policies in place concerning the respective roles of law enforcement and prosecutors in entering into and documenting informant agreements.

Constitutional rulings regarding documentation and disclosure of informant-related and informant-provided evidence. Constitutional requirements mandate, as a matter of due process, that material information regarding the credibility of a witness be disclosed to the defense at a criminal trial, but they do not mandate any particular set of practices either for the documentation or the disclosure of such evidence. In Giglio, the U.S. Supreme Court held that when “reliability of a given witness may well be determinative of guilt or innocence, nondisclosure of evidence affecting credibility falls within [the] general rule” that “suppression of material evidence justifies a new trial[,] irrespective of the good faith or bad faith of the prosecution.” Giglio v. United States, 405 U.S. 150, 153-154 (1972) (citing Brady v. Maryland, 373 U.S. 83, 87 (1963)) (internal quotation marks omitted). Most recently, in Wearry v. Cain, the Supreme Court found that concealed information undermining the credibility of an incentivized jailhouse informant witness, “[b]eyond doubt . . . suffice[d] to undermine confidence in Wearry’s conviction.” 577 U.S. 385 (2016). Further, while the outright fabrication of testimony by an informant would violate due process, such intentional lies may be quite difficult to prove. See Napue v. United States, 260 U.S. 265 (1959) (holding that it violates due process for the government to knowingly introduce fabricated testimony). The careful documentation of evidence can help to ensure that constitutional disclosure requirements are met, as well as help to detect any outright fabrication of evidence by an informant. Further, when these due-process rules do not clearly apply in cases that are resolved through a guilty plea, agency practices are a crucial safeguard. See, e.g., Markus Surratt, Incentivized Informants, Brady, Ruiz, and Wrongful Imprisonment: Requiring Pre-Plea Disclosure of Material Exculpatory Evidence, 93 Wash. L. Rev. 523, 531 (2018).

The Supreme Court has held that an informant’s identity may be kept confidential unless the identity of the informant is important to the defense case. In Roviaro v. United States, the Court recognized “the Government’s privilege to withhold from disclosure the identity of persons who furnish information of violations of law to officers charged with enforcement of the law.” 353 U.S. 53, 59 (1957). However, this “informer’s privilege” is qualified, and a court may override the privilege and order disclosure of the informant’s identity if disclosure is “relevant and helpful to the defense of an accused, or is essential to a fair determination of a cause.” Id. at 60-61.

The need for documentation of informant-related evidence. When constitutional rulings do not mandate any particular set of documentation regarding informant background, statements, or testimonial histories, it is all the more important that agencies do so. See Emily Jane Dodds, Note, I’ll Make You a Deal: How Repeat Informants Are Corrupting the Criminal Justice System and What To Do About It, Wm. & Mary L. Rev. 1067, 1080-1081 (2008) (explaining that open-file policies, which are not mandated by the U.S. Constitution, often omit prior testimonial history and that this information thus is often not passed on to the defense). Large-scale scandals have resulted in situations in which the same informant repeatedly provided false information in exchange for leniency. In one high-profile case, Lesley Vernon White admitted to having provided false testimony in dozens of cases in California in exchange for leniency and other benefits. Robert Reinhold, California Shaken over an Informer, N.Y. Times, Feb. 17, 1989. In other situations, information concerning the use of an informant was documented, but it was concealed. Such was the case in Orange County, California, in which prosecutors maintained a secret informant database that contained exculpatory information, the revelation of which led to civil-rights investigations. Dahlia Lithwick, You’re All Out: A Defense Attorney Uncovers a Brazen Scheme to Manipulate Evidence, and Prosecutors and Police Finally Get Caught, Slate (May 28, 2015). Some commentators have argued that exchange of complete information regarding prior testimony by an informant should be constitutionally mandated, due to the importance of such information in regard to potential impeachment of witnesses at trial. See Dodds, supra, at 1080-1081 (discussing a possible expansion of the Brady materiality requirement to include an informant’s prior testimonial history).

A special problem arises from the widely shared implicit understanding that informants will be rewarded for producing information, even when the government has not expressly promised them any benefit. Informants, especially experienced ones, often proactively collect information or fabricate evidence in the absence of an agreement with the government, and then collect rewards after the fact. The informants then can testify truthfully that they were not promised a benefit, which is misleading to juries that do not understand such implicit arrangements. Russell D. Covey, Suspect Evidence and Coalmine Canaries, 55 Am. Crim. L. Rev. 537, 573 (2018). The government also can assert truthfully that the informant was not their agent, thereby evading the constraints of Massiah v. United States, 377 U.S. 201 (1964), which prohibits the governmental use of informants to deliberately elicit evidence from charged suspects. Raeder, See No Evil, supra, at 1436. A rule that any use of an informant must be accompanied by an arrangement documented in writing can help to prevent such conduct.

In addition to documenting the identity of informants, this Section calls for the documentation of prior testimony by informants. Recantations are common among informants. Documented prior statements and testimony provide important impeachment or post-conviction evidence. “Murder cases most often involve recantations by supposed eyewitnesses, including co-defendants and the actual criminals, they also include a significant number of recantations by jailhouse snitches or other informants or witnesses who claim the defendant confessed to them.” Samuel R. Gross & Alexandra E. Gross, Nat’l Registry of Exonerations, Witness Recantation Study: Preliminary Findings, May 2013 6 (2013), available at https://repository.‌law.‌umich.edu/cgi/viewcontent.cgi?article=1090&context=other (emphasis added). One study found that “official misconduct is more common in recantation exonerations (73%) than in non-recantation exonerations (51%).” Id. at 9. “[A] very high percentage of recantation exonerations involve perjury or false accusations . . . .” Id. In the Gross and Gross study, “[a]lmost all of the recanting witnesses . . . lied in their accusations rather than making honest mistakes.” Id. For example, Cameron Todd Willingham was “executed in Texas for the 1991 deaths of his three young children, who perished in a house fire. Jailhouse informant Johnny Webb, who testified that Willingham confessed to him, later recant[ed] his testimony.” Katie Zavadski & Moiz Syed, 30 Years of Jailhouse Snitch Scandal, ProPublica (Dec. 4, 2019).

Types of documentation. The need to carefully document informant-provided evidence is particularly heightened given concerns about the reliability of such evidence. When feasible, documentation should include the recording of informants’ statements. “To enable the defendant to challenge the veracity of the witness effectively, and a jury to assess his credibility, all interviews with potential trial witnesses should be electronically recorded either by audio or videotaping.” Bennett L. Gershman, Witness Coaching by Prosecutors, 23 Cardozo L. Rev. 829, 861 (2002). Due to the concern that absent a recording it is far more difficult to corroborate statements by informants or investigate their reliability, scholars have recommended that agencies and jurisdictions move toward recording of informant-provided evidence. See, e.g., Robert Mosteller, The Special Threat of Informants to the Innocent Who Are Not Innocents: Producing “First Drafts,” Recording Incentives, and Taking a Fresh Look at the Evidence, 6 Ohio St. J. Crim. L. 519 (2009). An Oregon statute provides that custodial interviews conducted by law enforcement within law-enforcement agencies should be recorded electronically “if the interview is conducted in connection with” certain stated crimes, although there is an exception when “the state demonstrates good cause for the failure to electronically record the interview.” Or. Rev. Stat. § 133.400 (amended effective Jan. 2020). The statute enumerates what may be considered “good cause” and specifically addresses situations in which “[e]lectronically recording the custodial interview would jeopardize the safety of any person or the identity of a confidential informant.” Or. Rev. Stat. § 133.400(7)(b)(D). For purposes of the statute, law-enforcement facilities include “a courthouse, building or premises that is a place of operation for a municipal police department, county sheriff’s office or other law enforcement agency at which persons may be detained in connection with a juvenile delinquency petition or criminal charge.” Or. Rev. Stat. § 133.400(7)(c). If recording is not possible, then near-contemporaneous documentation is essential. For example, the Los Angeles Police Department requires that at least two officers be present when meeting with an informant and that both officers be capable of understanding the informant’s language. L.A. Police Dep’t, L.A. Police Department Manual: Policy § 544.30, available athttp://www.lapdonline.org/lapd_manual/volume_1.htm#544.

Although this Section calls for recording informant interviews and carefully documenting their use, different rules are needed to protect the identities of informants whose identities are at least in part confidential, particularly when the reason for doing so is out of a concern for informant or public safety. The New York Police Department provides in its patrol guide that police officers may not record conversations with informants who are considered confidential. Body-Worn Cameras: Frequently Asked Questions about Body Cameras: When are Officers Required to Record Video?, NYPD, https://www1.nyc.gov/site/nypd/about/about-nypd/equipment-tech/body-worn-cameras.page (last visited June 2, 2020) (“Officers may not record sensitive encounters, such as speaking with a confidential informant”). Although such a broad rule may not be necessary, agencies should set out procedures for ensuring confidentiality, when appropriate, while also ensuring that evidence is carefully documented.

Informant databases. As a best practice, agencies should track the use of all informants, including informants treated as confidential, and maintain that data over time. A range of individual agencies at the local, state, and federal level already do that. In the New York City Police Department, “once informants are approved for use they are photographed, fingerprinted and entered into a closely guarded registry. Any officer who deals with them is required to log the contact in the registry, with a record of any payments made.” Alan Feuer & Al Baker, Officers’ Arrests Put Spotlight on Police Use of Informants, N.Y. Times, Jan. 27, 2008. In its own database, “[t]he FBI tracks the productivity of [confidential informants] by aggregating their ‘statistical accomplishments,’ i.e., the number of indictments, search warrants, Title III applications, and other contributions to investigative objectives for which the [confidential informant] is credited.” U.S. Dep’t of Just., Off. Inspector Gen., The Federal Bureau of Investigation’s Compliance with the Attorney General’s Investigative Guidelines: Chapter Three: The Attorney General’s Guidelines Regarding the Use of Confidential Informants (2005). Such tracking systems “create[] a database of relevant information for prosecutors who can then evaluate the reliability of their witnesses and avoid wrongful convictions before they happen.” Alexandra Natapoff, The Shadowy World of Jailhouse Informants: Explained, Appeal (July 11, 2018), https://theappeal.org/the-shadowy-world-of-jailhouse-informants-an-explainer.

A database can facilitate implementation of documentation and disclosure requirements. Some jurisdictions already have adopted such database systems. In 2019, Connecticut enacted legislation that created “the nation’s first statewide system to track the use of jailhouse informants, including any benefits offered in exchange for their testimony.” Dave Collins, Lying Prisoners: New Laws Crack Down on Jailhouse Informants, Hartford Courant (Sept. 16, 2019). The statute provides for a series of jailhouse-informant-related disclosures, including: information about the criminal history of such witnesses; cooperation agreements and any benefits offered; all statements made; any past recantations of such testimony; and information about any prior testimony by the witness. Id. In 2020, the Oklahoma state legislature passed similar legislation creating a statewide database to track jailhouse informants. See Okla. Sen. Bill 1385 (effective Nov. 2020). That act also creates a reporting requirement, wherein “the District Attorneys Council shall publish an annual report of aggregate, de-identified data regarding the total number of cases tracked [in the database] . . . and the number of cases added during the previous fiscal year . . . .” Id.

Careful auditing of informant databases can help identify more systemic issues regarding the use of informants. The U.S. Department of Justice’s audit of U.S. Drug Enforcement Administration (DEA) practices found the DEA’s database of confidential sources to be “unreliable” and “incomplete.” U.S. Dep’t of Justice, Off. Inspector Gen., Audit Division, The Drug Enforcement Administration’s Payments to Confidential Sources: Executive Summary 6-7 (2005) (lacking descriptions of the justification of payments, for example). The Justice Department made 12 recommendations in its report, including that the DEA “add a module to an existing database system to track confidential source impeachment information” and “account for all payments made to a confidential source by the DEA, not just payments using DEA-appropriated funds.” Id. at 8. In a 2015 audit, the Justice Department again “found that the DEA did not adequately oversee payments to its sources, which exposes the DEA to an unacceptably increased potential for fraud, waste, and abuse.” U.S. Dep’t of Justice: Off. Inspector Gen, Audit of The Drug Enforcement Administration’s Management and Oversight of Its Confidential Source Program i (2016). The 2015 audit showed that “the DEA did not appropriately track all confidential-source activity; did not document proper justifications for all source payments; and, at times, did not adequately safeguard traveler information.” Id. at iii. The U.S. Constitution requires the government to disclose impeachment material regarding informants, namely, any information that would cast doubt on the person’s credibility. Many states specify exactly what that material must include, such as the informant’s previous statements, his or her criminal history, benefits received or promised, any testimony in prior cases, and recantations. Tracking systems help ensure that such information is fully collected and properly disclosed. Certain states require disclosure to the defense of past recantations by informants or witnesses. For example, Illinois requires the disclosure of “whether at any time the informant recanted that testimony or statement and, if so, the time and place of the recantation, the nature of the recantation, and the names of the persons who were present at the recantation.” 725 Ill. Comp. Stat. Ann. § 5/115-21(c)(5) (2019); see also Conn. Sen. Bill 1098 (effective Oct. 2019) (“Whether at any time the jailhouse witness recanted any testimony subject to the disclosure and, if so, the time and place of the recantation, the nature of the recantation and the name of any person present at the recantation”); Neb. Rev. Stat. § 29-4704(1) (2019) (“Any occasion known to the prosecutor in which the jailhouse informant recanted testimony about statements made by another suspect or defendant that were disclosed to the jailhouse informant and any transcript or copy of such recantation.”); Okla. Sen. Bill 1385 (effective Nov. 2020) (requiring disclosure of a range of information, including recantations by jailhouse-informant witnesses); see also Dodd v. State, 993 P.2d 778 (Okla. Crim. App. 2000) (instituting new rules for jailhouse-informant testimony that require prosecutors to, before trial, provide information including other cases in which the informant testified and any recantations).

Table of Contents