At a recent London School of Economics seminar, George Staple, director of the Serious Fraud Office, questioned whether a randomly selected jury of 12 members of the public was a suitable tribunal to try the complex criminal fraud cases of today's financial world.

This question has been posed for at least 10 years. As Staple reminded his audience, the Roskill Committee on fraud trials recommended in December 1995 that, in this category of case, the jury should be replaced by a “fraud trials tribunal” comprising a judge and two lay members “selected from a panel of persons who have skill and experience in business generally, and experience of complex business transactions”.

While accepting and subsequently implementing many of the committee's recommendations for the improvement of trial and pre-trial procedures and the modernisation of some of the more antiquated rules of evidence, the Government declined to adopt the proposal. Nor did the opposition parties show any enthusiasm for such a change. No doubt politicians of all parties were (rightly) reluctant to advocate the removal of jury powers and responsibilities from ordinary people.

The politicians may also have been impressed by the comments of one of the committee members, Walter Merricks (soon to be the Insurance Ombudsman), in his note of dissent from the majority conclusion. He said: “The submissions made to us by those who are closest to the existing system were overwhelmingly in favour of retaining the jury. The vast majority of the police, the solicitors' profession (from both the defence and prosecution perspectives), the magistrates and the Bar opposed the removal of jury trial.

“While views among the judiciary were divided, it is clear that many judges had grave reservations about removing the right to jury trial.”

He continued: “The submissions which did favour an alternative mode of trial came largely from the financial and accountancy world. When pressed in oral evidence it became clear that most of them based their views on generalised impressions.

“Among the judges and lawyers who gave evidence, none suggested that they had regularly come across cases in which the verdicts returned by juries in fraud cases indicated that they had not understood the evidence. On the contrary, many spoke of the dedication and application juries brought to their unfamiliar task and how their verdicts often reflected an apparently sophisticated evaluation of the charges and the evidence combined in a common sense result.”

Has the experience of those professionals in the criminal justice system changed dramatically over the past decade? Talk to judges as well as prosecution and defence lawyers and you will find it has not. A minority are now in favour of introducing an alternative to jury trial in complex fraud cases, but most judges and lawyers continue to be impressed by the way lay juries handle these demanding cases.

The problems begin when you consider the alternatives to jury trial. The only alternative that has attracted any real support is the Roskill Committee's fraud trials tribunals proposal – all other ideas, such as trial by judge alone, are widely regarded as unacceptable.

The Roskill proposal has two practical advantages. The first is that trials might be somewhat shorter, and therefore cheaper, for the public purse, which usually has to bear the considerable cost of these cases. However, it is doubtful whether the overall saving in cost would be as great as has been suggested. Much of the expense in large fraud cases comes at the investigation and preparation stages where there would be no saving. Indeed, it is possible that more time would be spent on preparation.

The second advantage is that a finding of guilty could be set out in a reasoned judgment which explained the precise factual basis on which the defendant was convicted and on which he could be sentenced. Where the charge is one of conspiracy to defraud, or one of the other broader offences prosecutors like to allege, a simple jury verdict of guilty or not guilty may leave a lot of important questions unanswered. However, this difficulty could often be removed by more precise framing of charges and/or by inviting the jury to return special verdicts in some cases.

But the practical advantages of the Roskill proposals must be set against the disadvantages of introducing a new criminal trial process for only one species of serious criminal conduct.

It is sometimes pointed out that a large proportion of criminal cases in the UK are not tried by juries but by stipendiary or lay magistrates. This is surely a red herring. Trial by magistrate is appropriate and convenient for most relatively minor offences which come before the courts. But nobody would suggest that magistrates should try allegations of serious crime and I have yet to come across anyone who thinks complex fraud cases are suitable for trial by magistrates.

The use of experienced business people to serve on fraud trials tribunals is superficially attractive but on closer examination it is fraught with problems. Such people are likely to bring preconceptions and assumptions to the case based on their experiences, which may be quite different from those of the person they are trying. They may, for example, find it impossible to believe that someone else could have made a genuine mistake when it is one they would have never made.

One of the strengths of the jury system is that randomly selected juries have proved exceptional at drawing the line between what is and what is not dishonest and at distinguishing the real rogues who deserve to be convicted from those whose conduct may have been foolish or irregular but do not deserve to be branded as criminals.

Serious doubts must exist as to whether the judgment of the semi-experts who the Roskill Committee would like to serve on fraud tribunals would be as reliable. They may set the standard too high through fear of appearing to encourage practices the business world or authorities consider undesirable.

Judging the truth or falsity of allegations of serious criminal conduct is a hugely important task. A wrongful conviction of an innocent man or woman is likely to have disastrous consequences for him or her. And a wrongful acquittal of a guilty person may have consequences for the community as a whole.

It is a sobering thought that in any legal system miscarriages of justice will occur. But to take complex and serious fraud cases away from juries, which have a long and successful record of tackling them, would only increase rather than reduce the risk of this happening.