+44 (0)20 7353 6000
01 Jun 2010

Corker Binning questions the AG’s guidelines on plea discussions in fraud cases

In an article published on Mondaq, Nicola Finnerty and Gemma Tombs of Corker Binning ask whether the AG’s guidelines on plea discussions in fraud cases are fit for purpose.

Plea Discussions in Fraud Cases: Are the AG’s guidelines fit for purpose?

The Attorney General‟s Guidelines on Plea Discussions in Cases of Serious or Complex Fraud were initially welcomed by most criminal practitioners, not least because it was anticipated that they would bring a degree of certainty for a defendant or suspect when deciding to plead guilty. However, one year on and with the judiciary having asserted its authority over how these agreements should work in practice, there seems little if any benefit to a defendant or suspect in even contemplating going down this route.


The guidelines came into being in May 2009 in response to a Fraud Review Recommendation in 2006 which called for a new framework for the conduct and acceptance of plea bargaining. According to Baroness Scotland, the guidelines were designed “to help prosecutors deal with fraud more effectively and efficiently to the benefit of the public and all parties involved” and the self-proclaimed purpose was „to narrow the issues in the case with a view to reaching a just outcome about acceptable pleas of guilty and preparing a joint submission as to sentence.‟

At the same time, the Consolidated Criminal Practice Direction (2002) was amended to provide further guidance on the procedure to be followed when embarking on plea discussions. They emphasised that the judge retains the absolute discretion to refuse to accept the plea agreement and that, fundamentally, “sentencing submissions should not include a specific sentence or agreed range other than the ranges set out in sentencing guidelines or authorities”.

Since then, it is understood that only a handful of cases have involved formal plea discussions. Of these, two have been the subject of stinging criticism by the judiciary, in R v Innospec and R v Dougall.

Inherent hazards in guidelines

These cases have highlighted the difficult (if not impossible) balance to be struck between meaningful plea discussions and the views of the judiciary. Importantly, they have exposed a number of potential hazards inherent under the guidelines which any individual or corporate considering this process will need to take into consideration.

  • Both parties must commit to written terms and conditions and confidentiality undertakings before any discussions are initiated. These undertakings seek to reassure the defendant that any information provided in the course of the discussions will not be disclosed to a third party other than for the purposes of the agreement. However, the guidelines are not primary legislation and therefore do not outweigh the disclosure obligations placed upon the prosecution by the CPIA 1996. Any information provided by the defendant could therefore find itself disclosed to co-defendants as unused material whether or not the plea discussions are successful.
  • The prosecution should also give an undertaking not to rely on any such information against the defendant should the discussions fail. In reality, this undertaking is of little effect. Firstly, in order to embark on plea discussions pre-charge, the suspect will have to participate fully and willingly in an interview under caution so that the prosecution can be „satisfied that the suspect‟s criminality is known‟. Clearly any admissions made once a suspect has been cautioned under PACE 1984 can be relied upon as evidence against the suspect. Secondly,
    and most significantly, if the discussions are successful the guidelines require the defendant to sign the plea agreement before it is put before the court. The defendant must therefore take the gamble of making a written confession as to his guilt before he even enters a court room and, importantly, without any certainty as to whether or not the court will even accept the agreement or dismiss it outright.
  • The most fundamental concern to a defendant will always be the sentence he or she receives and yet this is the aspect of plea discussions under the guidelines which is the least certain. Even if the court accepts a plea agreement, the judge is not bound to follow the basis of plea when passing sentence. Whilst the parties are entitled to put forward a joint sentencing submission, it can only go so far as to detail the appropriate sentencing authorities and applicable sentencing range in the relevant sentencing guideline. The recent cases of Innospec and Dougall have emphasised in the strongest terms that the merest hint of collusion between the parties as to sentence will be met with stern criticism and is likely to result in the court not accepting the agreement in its entirety.

The lack of control over sentence is recognised in so far as the guidelines envisage that a defendant may seek an advance sentence indication (pursuant to the guidance in R v Goodyear) before pleading guilty under the terms of a plea agreement. However, a judge is not bound to accede to any such application and, worryingly, the practice direction seems to envisage that before such an indication is given the defendant will have signed the plea agreement in any event (so as to make it binding on him in the event that indication is unfavourable).

Guidelines not fit for purpose

So is there any benefit to the individual in entering into the plea discussion process at all? Whilst there is no certainty at the end of what is an onerous process and, importantly, whilst there are no safeguards at all if the plea discussions fail or the judge refuses to accept the agreement, an individual may well find himself in a worse position by entering this process.

There is clearly a need for a robust plea negotiation framework but to be effective it must provide a sufficient incentive to the individual to plead guilty at the earliest opportunity. Arguably this can only be achieved through primary legislation and with the backing of the judiciary. Until then “existing practices” (which the guidelines specifically state should not be prevented) such as both sides agreeing a basis of plea and advance sentence indications may well be considered the safest and most effective course of action.

Nicola Finnerty is a partner and Gemma Tombs a solicitor at Corker Binning, the specialist fraud, regulatory
and criminal litigation firm. This article can also be found on Mondaq