by Douglas Ankney

“Cooperation is a horrible thing for clients. Doing law enforcement’s job and requiring someone to bargain for their freedom encourages an ugly, unfair, and unjust system to become even more so that way.”

Those are the words of an anonymous federal defense attorney who participated in a survey published as Why Criminal Defendants Cooperate: The Defense Attorney’s Perspective, 117 Nw. U. L. Rev. 1351 (2023) (“Survey”). This short review briefly examines the highlights of the Survey.

Part 1 describes the Survey’s background.

Part 2 looks at the of history, method, and current prevalence of cooperation agreements within the federal criminal justice system.

Part 3 examines the factors considered in the decision to cooperate, the rated importance of those factors, and the frequency of those factors influencing a decision to cooperate.

Part 4 considers what is required of defendants to obtain a cooperation agreement, defense attorneys’ perceptions of prosecutors’ conduct during the process, and why some defendants who desire to cooperate are rejected.

Part 5 reviews the manner in which cooperators are rewarded.

Part 6 discusses the training, or lack thereof, of the attorneys participating in the cooperation-agreement process.

Part 7 focuses on attorneys’ perception of the fairness of the process.

The Conclusion explores the implications of the Survey’s findings, suggests possible reforms in the practice of cooperation, and suggests further research in related areas.

Part 1: The Survey’s Background

The Survey’s authors are former Assistant U.S. Attorney Jessica A. Roth, Professor of Law and Co-Director at the Jacob Center for Ethics in the Practice of Law, Benjamin N. Cardozo School of Law, Yeshiva University; Anna D. Vaynman, Ph.D. Candidate in Psychology & Law, John Jay College of Criminal Justice, CUNY Graduate Center; and Steven D. Penrod, Distinguished Professor of Psychology, John Jay College of Criminal Justice, CUNY.

The Survey reported findings based on the anonymous responses of 146 defense attorneys to key questions surrounding the process of defendants entering into “cooperation agreements” within the federal criminal justice system.

The defense attorneys surveyed were federal public defenders, defense attorneys in private practice, and Criminal Justice Act (“CJA”) panel attorneys. (CJA panel attorneys are from private practice but are appointed to represent indigent defendants—often when the Federal Public Defender’s office has a conflict of interest in representing a particular defendant. 18 U.S.C. § 3006(a).) The Survey concentrated on attorneys practicing in three jurisdictions: (1) the Southern District of New York (“SDNY”) (48 Attorneys), (2) the Eastern District of Pennsylvania (“EDPA”) (22 Attorneys), and (3) the Eastern District of Virginia (“EDVA”) (39 Attorneys). However, ten of the Survey’s attorneys indicated that the majority of their practice was outside these jurisdictions (“Other”), while 27 respondents did not reveal their district of primary practice.

“It is a travesty that the government controls the entire process through judicial nonchalance and absurdly high sentencing guidelines.” (Another anonymous federal defense attorney)

Part 2: The History, Method, and Current Prevalence of Cooperation Agreements Within the Federal Criminal Justice System

“No practice is more ingrained in our criminal justice system than the practice of the government calling a witness who is an accessory to the crime for which the defendant is charged and having that witness testify under a plea bargain that promises him a reduced sentence.” United States v. Cervantes-Pacheco, 826 F.2d 310 (5th Cir. 1987). The Survey observed “[r]ewarding defendants for their assistance in the investigation and conviction of others is nothing new. Anglo-American courts have long upheld the practice of exchanging leniency for testimony. At its best, cooperation enables prosecutors to hold accountable the most culpable offenders, who would otherwise likely escape justice.” Indeed, a case from almost 150 years ago discusses the “ancient practice” of “approvement” wherein “an individual indicted for treason or felony could obtain a pardon in exchange for testifying truthfully against an accomplice upon the accomplice’s conviction.” The Whiskey Cases, 99 U.S. 594 (1879).

Allowing a defendant to snitch on his or her codefendants in exchange for leniency is justified on the dubious premise that the most culpable offenders at the top of the chain “rarely do the dirty work” and “[b]argaining for testimony is said to be necessary to detect and successfully prosecute certain crimes.” The Federalization of Organized Crime: Advantages of Federal Prosecution, John C. Jeffries Jr. and John Gleeson, 46 Hastings L. J. 1095 (1995) cf. Defending Substantial Assistance: An Old Prosecutor’s Meditation on Singleton, Sealed Case, and the Maxfield-Kramer Report, Frank O. Bowman III, 12 Fed. Sent’g Rep. 45 (1999).

Even though cooperation existed in America’s courts for centuries, it was not possible to effectively measure the practice until the United States Sentencing Guidelines (“U.S.S.G.” or “Guidelines”) went into effect in 1987. The Guidelines provided for fixed sentencing ranges that, when enacted, were mandatory and allowed little room for judges to depart from them. Nevertheless, U.S.S.G. § 5K1.1 “expressly authorized downward departures, upon government motion, where a defendant provided substantial assistance to the government in the investigation and prosecution of others.”

In United States v. Booker, 543 U.S. 220 (2005), the U.S. Supreme Court held that the Guidelines were unconstitutional to the extent that they were mandatory and that the Guidelines would be advisory going forward. While the courts are now free to sentence below the Guidelines range without a motion from the government, sentencing courts must provide explanation for any departure—Gall v. United States, 552 U.S. 38 (2007)—and U.S.S.G. § 5K1.1 is frequently cited to provide justification. Another mechanism often used in some circuits in cooperation agreements is Federal Rule of Criminal Procedure 35(b) (“Rule 35(b)”) that authorizes a judge to “reduce a previously imposed sentence on account of a defendant’s cooperation.” Cooperation agreements generally begin with defendants disclosing to prosecutors information thought to “be useful to the government. Prosecutors then evaluate the information and the defendant’s credibility and record and assess whether to proceed.” If the prosecutor decides to go forward, a cooperation agreement is offered with terms that usually include the defendant pleading guilty, providing truthful information, and substantially assisting the government in investigating and prosecuting others. In jurisdictions where U.S.S.G. § 5K1.1 departures are the primary mechanism of rewarding cooperators, sentencing is delayed until the defendant’s cooperation is complete. In jurisdictions where Rule 35(b) is the primary mechanism, defendants are sentenced before completing their cooperation. Upon completion, the government moves the District Court to adjust the initial sentence downward to reflect the cooperation.

In fiscal year 2021 (“FY 2021”) (covering the period October 1, 2020, to September 1, 2021), the SDNY reported 943 cases with individuals sentenced (“Cases”) with a guilty plea rate of 95.9% with § 5K1.1 departures in 12.1% of those Cases. In the EDPA, it was 446 Cases with a guilty plea rate of 93.7% and § 5K1.1 departures in 23.8% of those Cases. The EDVA, known for its “Rocket Docket,” reported 968 Cases, guilty plea rate of 96.4%, and § 5K1.1 departures in 4.1% of those Cases. However, the EDVA extensively employed Rule 35(b) reductions. (In FY 2021, there were 658 Rule 35(b) reductions in the entire federal system with 76 occurring in the EDVA.) When Rule 35(b) sentence reductions are added to § 5K1.1 reductions, the total in the EDVA for those Cases is 12%.

The Survey observed that throughout the federal criminal justice system, “approximately 9.6% of federal defendants” sentenced in FY 2021 “received an adjustment to their sentences based on substantial assistance to the government[.]” This equaled “approximately 5,493 individuals, who on average received over a 54.5% reduction in their sentence relative to their otherwise-applicable Guideline range.”

“The longer I practice the less I think cooperation makes sense for most of the defendants who are eligible to cooperate.” (Another anonymous federal defense attorney)

Part 3: The Decisive Factors, Their Rated Importance, and Their Frequency of Influence in a Decision to Cooperate.

The attorneys were asked to identify which type of case most frequently involved an opportunity to cooperate. The Survey’s authors then grouped the answers into three categories. The categories are listed below followed by the number of attorneys choosing that category: (1) Drugs—102, (2) Violent crimes—20, and (3) Nonviolent/white collar crimes not involving drugs—fraud, property offenses, theft/embezzlement, public corruption—16. (The authors omitted some responses because of overlap, e.g., some attorneys answered “firearms” without specifying if the offense was connected to a drug crime, a crime of violence, or a standalone offense.)

The attorneys were also asked to consider 27 factors (later combined for a total of 14, e.g., “defendant gender,” “defendant race,” and “defendant education level” were combined as “defendant demographic characteristics,” etc.). The attorneys were asked to rate the importance of each factor in the decision-making process on whether to enter into a cooperation agreement and to rate the frequency with which the factor arose. The ratings scale was 1 (very unimportant) to 9 (very important). The attorneys’ responses were then averaged together. The results, taken from Table 3: Mean Ratings of Importance and Frequency of Factors in Cooperation Decision of the Survey are reproduced below.

The number in the first set of parentheses is the “Importance” average, and the number in the second set of parentheses is the “Frequency” average:

Whether a statutory mandatory minimum sentence applies (8.39) (7.88);

Anticipated sentencing range without cooperation (8.33) (8.23);

Anticipated sentencing benefit (7.59) (6.38);

Trust in [the defense] lawyer (6.92) (6.67);

Defendant characteristics with potential impact on sentence and deportation (5.75) (4.26);

Type of cooperation required (5.65) (5.50);

Cultural, family, and safety concerns (5.54) (4.28);

Trust in prosecutor and judge (5.4) (4.50);

Concern about exposure for uncharged criminal conduct (5.39) (5.29);

Bail status (5.07) (4.07);

Pressure from defense attorney and defendant’s circle (4.14) (3.28);

Demographic characteristics (3.16) (2.34);

Remorse (2.77) (2.80);

Financial incentives (2.21) (2.41).

The findings confirmed that the primary reason for cooperation is related to sentencing. The overall sentence length and the perceived possibility of reducing that sentence is the motivation and the overriding concern. Remorse was rated at the bottom—more important than only financial incentives.

The authors reported that these answers were fairly uniform across all three districts, except for the factor of trust in the defense attorney. Attorneys from the SDNY rated this as significantly more important (7.49) than did attorneys from the EDVA (6.18).

The authors offered as a possible explanation “SDNY’s practice of requiring defendants, as a condition of cooperating, to plead guilty to additional crimes that were not initially charged. This practice, which is not the norm in the other two districts—and is required only by local custom rather than any positive law—increases the level of risk associated with cooperation. Defendants who pursue cooperation in SDNY take the usual risks associated with cooperation—including that their attempt will ‘backfire’ by failing to earn them a cooperation agreement and simultaneously rendering the odds of a successful defense of the case ‘remote.’ They also take the risk, even if they succeed in obtaining a cooperation agreement, that doing so will increase their sentencing exposure—elevating the baseline from which the judge will depart—with no guarantee that the government will make motion for a downward departure or, if they do, what the extent of any such departure will be.” Confidence in one’s lawyer in those circumstances is paramount.  

The above explanation may also impact another observation. The attorneys were asked to rate the frequency with which they broached the subject of cooperation with their clients on a scale of 1 (never) to 9 (always). The combined average was 7.94. Yet, the attorneys “reported that in 49.98% of the cases in which they discussed cooperation, the client declined to cooperate with prosecutors.” However, when broken down by district, defendants in the SDNY refused to cooperate at a much higher rate (60.27%) than defendants in the EDVA (35.58%) and defendants in the EDPA (53.04%).

The authors were “not aware of any prior published accounts establishing that the other two districts in [their] study do not similarly impose such a requirement.” Yet, apparently, even if the EDPA or the EDVA impose the requirement, it’s not as all encompassing, widespread, or uniform as in the SDNY.

In the SDNY, prosecutors expect potential cooperators “to tell the government about all of their criminal conduct throughout their lifetime as a precondition to a cooperation agreement” and require cooperators to “plead guilty to serious conduct that they reveal to the government.” Cooperation with Federal Prosecutors: Experiences of Truth Telling and Embellishment, Ellen Yaroshefsky, 68 Fordham L. Rev. 917 (1999). The practice in the SDNY “is believed (at least by prosecutors) to enhance cooperators’ reliability and credibility by putting their full criminal past before the judge and jury, demonstrating that they accept full responsibility for their crimes and have not withheld anything from prosecutors. It also increases the stakes for cooperators if they do not comply with the terms of their cooperation agreements, including the requirement that they provide only truthful testimony on the stand.”

“The practice of cooperation encourages exaggeration, misrepresentations and lying because of the substantial sentences clients face, especially in drug cases.” (Another anonymous federal defense attorney)

Part 4: The Requirements Placed on Defendants to Obtain a Cooperation Agreement, Perceptions of Prosecutors’ Conduct During the Process, and Why Some Defendants Who Desire to Cooperate Are Rejected

The Survey reported “[a]mong the 50.02% of clients who did express an interest in cooperating, most of them—83.7%—were able to successfully secure a cooperation agreement. A cooperation agreement was not secured in only 16.30% of the cases in which clients tried to cooperate. Thus, if a client did not receive a cooperation agreement, it was more likely because that individual chose not to pursue cooperation than because the prosecutor did not offer a cooperation agreement.”

The attorneys reported that “in 86.80% of their cooperation cases, defendants had to testify or be prepared to testify against others if necessary. In 80.71% of their cooperation cases, defendants were required to disclose uncharged criminal conduct in the proffer session. In 43.34% of their cooperation cases, defendants were required to plead guilty to previously uncharged criminal conduct. And in 8.73% of their cooperation cases, defendants had to wear a wire or other recording device.”

[Note: Noticeably absent from the Survey is any reference to the percentage of cooperators in drug cases being required to make “controlled buys” from their dealers or the percentage of cooperators required to initiate conversations with other defendants about purchasing large quantities of narcotics, resulting in charges of conspiracy to possess with intent to distribute being brought against those other defendants.]

The defense attorneys also reported several common observations of the prosecutors’ conduct during the process of entering into a cooperation agreement. These are listed below in correlation to the type of offense (taken from the Survey’s Table 6) followed by correlation to district (taken from the Survey’s Table 7):

Prosecutors were open to fact and/or charge bargaining—by offense, (Drugs 34.83%) (Violent Crimes 19.56%) (Nonviolent Crimes 58.27%); by district, (SDNY 21.14%) (EDPA 26.89%) (EDVA 53.88%) (Other 47.50%);

Prosecutors were willing to set limits on the extent of the client’s required cooperation—by offense, (Drugs 27.38%) (Violent Crimes 19.33%) (Nonviolent Crimes 29.18%); by district, (SDNY 16.59%) (EDPA 28.06%) (EDVA 33.29%) (Other 49.00%);

Prosecutors initiated cooperation agreements—by offense, (Drugs 64.96%) (Violent Crimes 44.33%) (Nonviolent Crimes 55.36%); by district, (SDNY 58.50%) (EDPA 73.68%) (EDVA 58.84%) (Other 60.90%);

Prosecutors made specific sentencing recommendations to the judge—by offense, (Drugs 40.83%) (Violent Crimes 5.38%) (Nonviolent Crimes 35.67%); by district, (SDNY 15.06%) (EDPA 35.74%) (EDVA 58.18%) (Other 38.71%);

Prosecutors told the judge the type of assistance provided by the client—by offense, (Drugs 90.23%) (Violent Crimes 95.00%) (Nonviolent Crimes 95.71%); by district, (SDNY 99.17%) (EDPA 96.00%) (EDVA 75.74%) (Other 95.00%); and

Prosecutors told the judge if the client pleaded guilty to additional crimes that were not initially charged—by offense, (Drugs 65.00%) (Violent Crimes 87.44%) (Nonviolent Crimes 71.27%); by district (SDNY 90.61%) (EDPA 76.00%) (EDVA 17.88%) (Other 88.33%).

Some notable observations from the above findings: (1) prosecutors are significantly more open to fact/charge bargaining in nonviolent crimes (58.27%) than drug crimes (34.83%) and violent crimes (19.56%); (2) prosecutors in the EDVA are more willing to make specific sentencing recommendations to the judge (58.18%) than prosecutors in the SDNY (15.06%) or those in the EDPA (35.74%); and (3) prosecutors in the EDVA are much less likely to inform the judge that cooperators pleaded guilty to uncharged offenses (17.88%) than prosecutors in the SDNY (90.61%) or the EDPA (76.00%). However, this last observation may be due to the fact that cooperators in the EDVA are infrequently required to plead guilty to uncharged offenses.

Attorneys also reported “confirmation bias” on the part of prosecutors. That is, prosecutors tended to believe the first story they were told and rejected later versions by other clients, even where the later version was true. [Editor’s note: This is an extremely insightful and important point and likely helps explain and can serve as a fairly reliable predictor of a significant number of wrongful convictions.]

Of the many reasons defendants were unable to secure a cooperation agreement, the two most frequently cited were issues related to the information provided by the defendant: “either the information was deemed by the prosecutor to be insufficient or unhelpful or the defendant lied.”

The below list of reasons defendants were unable to secure a cooperation agreement is taken from Table 5 of the Survey with the cited frequency in parentheses:

Defendant’s information was insufficient, unhelpful, or contradictory to prosecutor’s evidence (50);

Defendant lied, did not tell the whole story, or minimized (20);

Prosecutor did not believe defendant (10);

Prosecutor was unfair, lazy, untruthful, or had all the power (9);

Defendant committed new crimes, feared exposure for uncovered crimes, or violated terms of cooperation agreement (5);

Defendant was not interested in cooperation or did not sufficiently benefit from it (5);

Defendant was fearful of safety and stigma of consequences or refused for ethical reasons (4);

Defendant did not trust law enforcement or defense attorney or didn’t listen to defense attorney’s advice (3);

Defendant was too late to cooperate (3);

Defendant was too culpable relative to others to cooperate (3); and

Politics within/among districts (2).

“To me, cooperation allowed me to come clean about my crimes and conduct and was an important part of my making amends on a number of levels.” (Richard Bistrong, former cooperator with federal prosecutors for five years)

Part 5: Manner of Rewarding Cooperators

The responding attorneys “reported that 68.72% of their cooperation cases involved a section 5K1.1 motion or 18 U.S.C. section 3553(e) motion, 36.49% involved Rule 35(b) motions, 14.42% involved a non-prosecution agreement, 6.66% involved a deferred prosecution agreement, and 8.20% involved an informal cooperation agreement.” As discussed above, there were significant differences in method of rewarding cooperators across the districts. Attorneys practicing in the EDVA reported 64.95% of their cases involved Rule 35(b) motions compared with the SDNY’s 13.55%, the EDPA’s 13.57%, and attorneys from Other Districts (11.00%). Conversely, the EDVA reported the lowest percentage of § 5K1.1 motions (23.10%) compared with the SDNY’s 82.54%, the EDPA’s 94.45%, and attorneys from Other Districts reporting 84.10%.

Additionally, when contrasting the types of crimes, nonviolent crimes again received favored treatment. Clients charged with nonviolent crimes received deferred prosecution in 13.67% of the attorneys’ cases; whereas, attorneys representing clients charged with violent crimes reported deferred prosecution in 9.88% of cases and drug crimes at a low of 4.45%. Similarly, non-prosecution agreements went to nonviolent-crime clients in 18.6% of cases, violent-crime clients in 8.45% of cases, and clients charged with drug crimes in 14.84% of cases. As explained above, cooperators who were sentenced to prison received a sentence, on average, of less than half (45.5%) of the sentence called for in their Guidelines range. However, the Survey is unclear as to whether this is 45.5% of the Guidelines range for the initial charged offense(s) or if it is 45.5% of the Guidelines range as calculated after cooperators pleaded guilty to all uncharged offenses committed over the course of their lifetime.

“Client cooperation is a despicable practice which is not a search for truth, but a search for convictions. Once my clients become cooperators, AUSAs all of a sudden are sympathetic to the difficult lives my clients have led.” (Another anonymous federal defense attorney)

Part 6: Training of Attorneys to Assist Clients Seeking Cooperation Agreements

Surprisingly, fewer than half of the attorneys surveyed reported receiving training in the process of assisting clients in obtaining a cooperation agreement. Just 43.2% indicated that they had received training, while 56.8% reported they had not. By district, the attorneys reported receiving training as follows: SDNY (40.4%), EDPA (57.1%), EDVA (36.8%), Other (50.0%). (Taken from Survey, Table 10) The difference in training, or lack thereof, was most pronounced based on the role of the attorneys with just 37.6% of CJA attorneys receiving training versus 78.6% of public defenders.

The types of training and number of participants, taken from Table 11 of the Survey, were:

CLEs [“Continuing Legal Education Classes”] – 25;

Mentorship and on-the-job-training – 18;

Seminars or conferences – 7;

Internal office training – 7;

Research – 3;

Other – 1 (One attorney wrote “during tenure as AUSA” but did not detail the type of training.)

As for those attorneys who received no training, they were asked what shaped their cooperation practices. The answers, and the number of attorneys giving a particular answer, were taken from the Survey’s Table 12 and reproduced as follows:

Experience – 51;

Mentorship – 28;

Best interests of the client – 6;

Personal beliefs and abilities – 5;

Observation – 3;

Research – 2;

Local policy – 1;

Other – 1 (One attorney wrote: “The two-door rule: you win on every count or you agree to cooperate.”)

“The #1 thing that makes the cooperation process unfair is mandatory minimum sentences—especially with drug cases. Because cooperation is often the only way a client can avoid the mandatory minimum sentence, the prosecution has all the leverage and sole discretion.” (Another anonymous federal defense attorney)

Part 7: Attorneys’ Perception of Fairness in the Cooperation Process

The authors of the Survey reported “[p]articipants were asked to rate on a scale of 1 to 9 their agreement with statements designed to elicit their views on the fairness of cooperation. The first statement was ‘defendants who cooperate generally fare better than those who do not,’ which received an average rating of 8.21. The second statement was ‘cooperation agreements are the product of a fair process,’ with an average rating of 3.17. Such a low average indicates that federal defense attorneys who participated in this study felt that cooperation agreements are not the product of a fair process. And the third statement was ‘I believe it is my professional obligation to advise clients that they have an opportunity to cooperate,’ which received an average rating of 8.82.”

The authors also invited the attorneys to share open-ended comments. In addition to the anonymous quotes provided above, the attorneys commented on other aspects of the process. For example, the attorneys pointed out that it’s entirely up to the discretion of the prosecutor to determine if what the defendant provided was actually “cooperation.” There is no set standard or criteria as to what qualifies as cooperation. And the attorneys complained about the extent of the discretion given to prosecutors in the filing of § 5K1.1 or Rule 35(b) motions. Another oft-repeated theme focused on “the difficulty of getting prosecutors to follow up with such motions or other assistance with cooperating clients.” One attorney wrote: “There are just some prosecutors who don’t follow through. Once they get what they want from the client, they forget about them and often won’t even answer phone calls or emails.” Another stated that prosecutors “love getting the cooperation, but getting them to follow through with a motion in a timely manner is a real problem.”

“Several respondents discussed the lack of safeguards against untrue or distorted information working its way into the process. For example, one respondent noted that prosecutors ‘never double check or investigate the witnesses’ statements or allegations’ such that ‘people who are familiar with the system are able to manipulate the prosecutors.’” Another attorney stated that the practice of cooperation “encourages and rewards dishonesty.”

A perceptive attorney wrote “the cuts for cooperation are too large and cause judges to inflate initial sentences when they expect to reduce later for cooperation.” Another respondent stated: “Cooperation feels awful. You are helping your client but in the back of your mind you know you are also helping to prosecute someone else.”

Conclusion

If the criminal justice system of the United States were analogized with a human body wracked with cancer, the practice of defendants obtaining cooperation agreements with the government would be that place of the gaping wound. The authors of this Survey noted: “A growing body of plea-bargaining literature suggests that where the discrepancy between the sentence offered in a plea is so astoundingly more lenient than the sentence exposure at trial, anyone—guilty or innocent—will accept the plea. Plea Bargaining, Jenia I. Turner, 3 Reforming Criminal Justice: Trial and Pre-Trial Processes 73 (2017). Since the conditions of the places of confinement are often so astoundingly horrid (and the sentences faced so torturously long) that innocent people will plead guilty to escape the surreal nightmare of facing criminal charges—what will a defendant in such a situation say/do to put the blame on a codefendant and escape that same predicament? 

In Table 3 of the Survey, as discussed in Part 3 above, the top three factors influencing the decision to cooperate are directly related to the sentence faced by the cooperator and the anticipated reduction for his or her cooperation. “Remorse” was rated next to last. In a truly “just system,” there can be no room for conditioning leniency on a requirement that one guilty person testifies against another. This is all the more apparent when the person desiring leniency seeks out the prosecution with the proposition of assisting in the prosecution of another. As one of the attorneys said, such a practice “encourages and rewards dishonesty.” Desperation does not promote justice. The practice of rewarding criminal defendants for assisting the government in the prosecution of others must cease.

That being the case, it seems inappropriate to even suggest possible means of reform. Is a rattlesnake with one fang less deadly? Yet, it seems unlikely the practice of rewarding criminals for their assistance in prosecuting others is likely to end anytime soon. Attempts must be made to minimize the carnage and ruined lives.

The contours of the cooperation agreements should be spelled out with the same specificity as plea agreements. The details should include exactly what the cooperator must perform and exactly what the cooperator will receive in return. With regard to the filing of § 5K1.1 or Rule 35(b) motions, there must be definite timelines within the cooperation agreements—along with stiff penalties for failing to abide by those timelines (such as dismissal of charges as in speedy trial statutes).

Critically, attorneys involved in assisting defendants with cooperation agreements must be properly trained. In addition to adequate training, defense attorneys need to be involved in crafting the cooperation agreements in similar fashion as their involvement in drafting plea agreements or jury instructions. Where is it written that the power in these agreements must reside exclusively with the prosecutor? The amount of the incentive for cooperation needs to be examined. Reductions of more than 50% of defendant’s sentence seems to almost “beckon” cooperators to fabricate and misrepresent facts implicating codefendants in order to secure an agreement. The subject of mandatory minimum sentences continues to be debated. From the government’s position, mandatory minimum sentences are a “positive” in the context of cooperation as those sentences greatly motivate defendants to cooperate and assist in successful prosecutions of others. But at what cost to truth and integrity (and all too often, the factually innocent)?

The Survey has many limitations. Most noticeable of these is that no prisoners were surveyed. In seeking to know and understand the factors that most influence defendants to cooperate, it would seem the responses of the people who actually made the ultimate decision to cooperate are top priority.

There are other ethical considerations that need to be explored and addressed. For example, do prosecutors sometimes require cooperators to make controlled drug buys of “larger than usual quantities” simply to increase the criminal penalty to which the non-cooperating defendant will be exposed? Is that “creating a crime that otherwise would not have occurred?” Do prosecutors require cooperators to initiate and secure “conspiracies” with other defendants that otherwise would not have occurred?

Finally, further research is needed to shed light on the frequency of cooperators giving false or misleading testimony that results in convictions of the innocent. Is it substantially similar to that of jailhouse informants?  

Source: Why Criminal Defendants Cooperate: The Defense Attorneys Perspective, 117 Nw. U. L. Rev. 1351 (2023)

Subscribe today

Already a subscriber? Login