The SFO, with a budget of pounds l7.3m and 150 staff, investigates and prosecutes the headline-grabbing cases, but the bulk of fraud trials are handled by the Crown Prosecution Service's fraud division, which has about 350 serious and complex cases on its books at any time, many involving multimillion- pound sums.
The fraud division has a budget of about pounds l2m and 89 staff - 30 lawyers and seven accountants, plus support staff - which is split between four branches. Three are based at the London headquarters, two responsible for the capital and one for the south. The fourth branch is in York. Just over half of the cases are generated outside London.
Concern about the SFO's record in dealing with major frauds had led to two government reviews and the possibility of a merger with the CPS. That period of uncertainty ended last spring, when the Attorney General, Sir Nicholas Lyell, announced that the SFO, set up in 1988, would remain an independent organisation.
In the revised criteria sent out to law enforcement agencies in May last year, the key element in deciding whether the SFO should accept a case is if the suspected fraud "is such that the direction of the investigation should be in the hands of those responsible for the prosecution". Yet some prosecutors question whether this has proved a more effective way of handling complex frauds. As one prosecutor comments: "It is something of an anomaly that the CPS was created to divide investigation and prosecution, and yet within five minutes the virtue appeared to be in combining the two."
According to research carried out for one of the government reviews of the SFO, the percentage of all defendants being convicted since early 1990 was 64 per cent for the SFO and 73 per cent for the CPS fraud division, although there were differences in their caseloads. Robin Booth, the head of the CPS fraud division, says: "Does it weaken our ability to deal with cases because we are not a fully multi-disciplinary body under one roof? No, I don't think it does. In some ways we actually gain from the divide between prosecutor and investigator because it removes any problems of blurred responsibilities and accountability."
The fraud division came under fire in March last year when a multimillion- pound mortgage fraud trial was thrown out by the judge at the end of a prosecution case that had taken six months. After defence submissions that the prosecution case had become "oppressive and unmanageable", Judge Thomas Crowther discharged the jury at Newport Crown Court, Gwent, to avoid an "expensive disaster". The judge said: "A jury determines guilt or innocence on the evidence that they can comprehend or remember. I cannot conscientiously say that I can make that assumption knowing that the jury is likely to retire to consider its verdict in July or August."
One of the defence solicitors, Tim Greene, of Birnberg & Co, says the difficulties were caused by the prosecution "chucking everything in instead of pinpointing a particular aspect and ditching the rest.
"The prosecution took nearly as long as the whole Maxwell trial, and the idea that the Maxwell finances were easier to disentangle than the purchase of two buildings is ridiculous."
The Newport case started with a 37-count indictment and 15 defendants, which the prosecutors envisaged would be split into about six trials. In the event, some defendants pleaded guilty and there were only three trials. One defendant absconded and seven were acquitted. Seven others either pleaded guilty or were convicted at trial, receiving sentences ranging from three and a half years to conditional discharge. The trial stopped by the judge involved seven defendants facing a four-count indictment. One prosecutor who was involved said: "We did not fail - at worst it was a draw."
The case renewed calls for juries to be replaced by expert panels headed by a judge. A few months later, it was learnt that the Home Secretary, Michael Howard, had asked officials to take soundings from senior legal figures on whether juries should go in order to save time and money. Inquiries among crown prosecutors dealing with serious fraud, however, found that most were happy with juries and were not aware of any cases over the past few years that had been lost because of perceived jury problems.
One of the fraud division's cases has the dubious distinction of appearing in the Guinness Book of Records as the longest single fraud trial. The Britannia Park case, involving the collapse of a theme park near Reanor, Derbyshire, in 1985, began on 10 September 1990 and ended on 4 February 1992, notching up 252 working days. At the subsequent appeal in July 1994, the Court of Appeal considered whether the length of the trial had created a situation whereby a fair trial was impossible. But it found that the jury had shown no difficulties in understanding the evidence. However, Lord Justice Farquharson, sitting with Mr Justice Gatehouse and Mr Justice Scott Baker, said the nine-day appeal had made it clear that "no trial by jury should be permitted to last this length of time. The risk of an accused not getting a fair trial because of the pressure upon judge or jury is too great."
Lord Justice Farquharson also said that prosecutors had to be reconciled to the fact that their case might be weakened by being broken up into separate trials, but added: "That is the price that must be paid to avoid the situation that arose in the present case."
While there will, no doubt, be much agonising about the future management of fraud trials, Mr Booth says: "I don't think it is out of control. Cases are now being prosecuted by the CPS and SFO which would not even have been investigated pre-1988. I don't think it is a terrible story of failure at all.
"Undoubtedly, there are improvements that could be made in the management of fraud trials and in the training of those involved. These need to be considered whatever decisions are taken on the use of juries in fraud trials.''
How the prosecutors compare
Serious Fraud CPS
Office Staff 150 89
Budget pounds 17.3m pounds 12m
No of cases 65 350
Success rate 64% 73%
Severity of sentences for those convicted are roughly similar. The key criterion for an SFO investigation is the need to have the investigation and the prosecution handled by the same body. Other factors taken into consideration include the belief that the fraud involves at least pounds 1m and issues of public interest.Reuse content