I have previously written on joint defense agreements ("JDAs") (in reverse chronological order):
- On Trump, Manafort and Joint Defense Agreements (Federal Tax Crimes Blog 9/14/18; 9/15/18), here.
- More on Joint Defense Agreements (Federal Tax Crimes Blog 5/15/18), here.
- On Joint Defense Agreements (Federal Tax Crimes Blog 11/23/17), here.
A very good discussion of JDAs in the special counsel's investigation appeared today, addressed to the general public: Darren Samuelsohn,
Trump team’s contact with Mueller targets could taint findings (Politico 10/8/18),
here. The author discusses the Trump legal team's use of JDAs and whether the use may have goals beyond those normally contemplated by JDAs. I urge readers who are interested in the intersection of politics and the criminal law to read the article because I believe it is quite good. (Full disclosure, I am quoted in the article.)
JAT Comments:
1. In my experience, a participant in a JDA withdraws at least by the time the participant agrees to cooperate and tell all. Of course, that participant will not be able to disclose items that the participant learned earlier from others in the JDA which are within the scope of the JDA. And, if the participant disclosed information to the other participants that is confidential to himself within the scope of the JDA, the participant can still disclose that information to the prosecutors pursuant to the cooperation agreement. The only prohibition on the cooperating participant is that he cannot disclose confidential information that the other participants in the JDA disclosed to him within the scope of the JDA. So, to use the stark example discussed in the article, if Trump or Trump's lawyers disclosed to Manafort or Manfort's lawyers Trump's admission to a crime within the scope of the JDA, Manafort could not disclose that admission to the prosecutors and, if Manafort did disclose the admission in violation of the JDA, the prosecutors would not be able to use the admission, directly or indirectly, in criminally prosecuting Trump. Other information that Manafort or his lawyers know that was not disclosed by Trump or his lawyers pursuant to the JDA can be disclosed to the prosecutor and can be used to prosecute Trump.
2. The problem comes in separating what the witness, here Manafort, knows independent of the confidential information he received pursuant to the JDA. If the prosecutor wants to prosecute Trump for a crime and uses information from a person in a JDA (here that would be Manafort), the prosecutor will have to prove that the prosecution is not based on any Trump confidential information that Manafort learned from Trump or his lawyers under the JDA. Unless the parties in the criminal action can agree, that would require a
Kastigar-like hearing where the prosecutor would have to prove that the case does not rely on that "tainted" information, directly or indirectly. That can, in many cases, be an impossible burden and would require suppression of any possibly tainted evidence or even, if so intertwined with the prosecution, dismissal of the case.
3. For this reason, JDAs can sometimes limit the benefit that prosecutors can get from cooperation from participants in a JDA and make prosecutors less willing to strike a deal with the participant if it appears that the evidence the participant has is tainted or potentially tainted. This requires delicate negotiations in reaching a cooperation agreement (usually by plea). Accordingly, in complex multi-target investigations, the marginal or less important actors may want to either not join a JDA or limit and carefully document what is received under the JDA so that that participant can maintain maximum flexibility in cooperation/plea negotiations.