Coroners and Justice Bill

That is an important point. In cases of domestic murder, for example, the chances of reoffending are usually very low, which will skew the statistics that the Secretary of State used. We know that for the type of offender that the Secretary of State was talking about—those with chaotic lives who sometimes find that they are sent to prison because the courts lose patience and cannot think of anything else to do with them—drug and alcohol treatment programmes are more effective, although not always effective, when delivered in the community than they are when delivered in prison. The reasons are obvious. Alcohol treatment, for example, is clearly the best way of dealing with violent offenders, even though some would say that all violent offenders must always be sent to prison. Alcohol and drug treatment programmes are far cheaper when delivered in the community than they are when delivered in prison. We could run more of them, and therefore prevent more crime, if we allocated our resources properly. We know that what works best for those whom we have to put in prison—the Justice Secretary is right: there are some people whom we have to put in prison—in reducing reoffending are training, education and work programmes. In fact, it turns out that those programmes are cheaper and more effective in prison than they are outside it, for various reasons. That tells us the direction in which prison policy should go, not just the direction in which sentencing should go. All I am saying is that we need a dynamic system in which whatever we know from the results of research is built into the process of drawing up the sentencing guidelines. That is all that amendment 20 aims to achieve. One issue in this debate is costs and the extent to which they should be a relevant consideration in criminal justice policy. Let me make it plain that I have absolutely no objection to considering the relative costs of sentences in deciding what should be in the guidelines. It seems obvious that if two sentences are equally effective in reducing reoffending, we should use the cheaper one, because we can do more of it and prevent more crime in the long term. However, I hope that the Government will make it clear that that is the not the same as making sentences in individual cases respond solely to cost, because not only individual cases would be affected. In fact, given the disparate nature of sentencing, doing that would probably not be cost-effective, either. It is also important to distinguish between the total resources available to the criminal justice system and the relative costs of different sentences. Sentencing guidelines should take into account relative costs, but that does not mean that they should take into account the total resources available to the system. That is the Government's business. It is for the Government to ensure that the resources are available to make sentencing in the criminal justice system work. As I understand it, that is the spirit of amendment 44, which has been tabled by Government Back Benchers.