Since its inception, more than three quarters of the cases prosecuted by the Serious Fraud Office have resulted in the conviction of at least one defendant, usually the principal offender. The conviction rate for all defendants over that period is 62 per cent.
If that sounds very familiar, it is merely what I said a few minutes ago, and it bears repetition.
The Director of the SFO has negotiated guidelines that govern the conduct of cases between himself and the regulators. If cases are of high-level public concern and display serious dishonesty on the part of individuals, they will be prosecuted, make no mistake about that. Regulatory action may be appropriate if the offences are technical or where urgent action is needed and it can best be taken by the regulator.
Impressive as those figures are, does my hon. and learned Friend agree that in his 62 per cent. he should include Mr. Levitt, who got off with a rather light sentence? Does he not agree that some of the cases have shown that serious fraud is too serious, too complex and too complicated to be dealt with by juries, and that the Government should follow the recommendations of the Roskill inquiry?
I believe that there is still much to be done to perfect our systems of trial, with new and improved information technology, which is currently being used in some cases, before we are driven to the stage of saying that such cases cannot be tried by jury and that they must be tried, as Roskill recommended, by some other tribunal. So far as the Levitt case is concerned, without dealing with any particular case, my right hon. and learned Friend now has power in appropriate cases to refer unduly lenient sentences in serious or complex frauds to the Court of Appeal for review.