Yes you need to be a little careful when taking statistics at face value.
Eddie’s 75% may well be correct (I haven’t checked). But, as has been mentioned, it should also be considered whether those in that 75% ever worked before they went inside. Also to be borne in mind is that, in the UK at least, one has to have committed either a single serious offence or quite a large number of less serious offences. Those going to prison in the latter category will almost certainly have already have been through the full graduation process of Conditional Discharges, Fines, Community Orders and Suspended Prison Sentences. The last two of these are almost always accompanied by considerable intervention and support from the probation service (with which I believe Eddie is familiar).
So to suggest that “prison doesn’t work” is a bit simplistic. It certainly works in that it affords the rest of us a brief respite from the activities of those incarcerated. It does not often work to rehabilitate a sizeable number of those sentenced to custody. But along with this statement has also to be asked the question “But would anything do so?” The probable answer is “No”. Large numbers of people sent to prison have shown themselves to be unresponsive to other non-custodial disposals and it could be argued that they are simply inclined to commit criminal offences whatever assistance they might be given to make them lead a non-criminal life. If they are not so inclined they will rehabilitate themselves as the people Mr Timpson employs illustrate.
When I was young (before the introduction of the Rehabilitation of Offenders Act) it was well known that a criminal record was likely to have an influence on your life for a very long time. In some respects it would affect you for life. It was a very powerful deterrent from committing criminal acts. The ROA sees offences “spent” – sometimes in a ludicrously short time and so that deterrent is no longer with us. I would argue that, certainly as far as deterring people from committing crimes, the ROA has actually increased the chances, particularly among young people, of criminality.
It is quite clear that the individual referred to in this question has not been influenced at all by a spell inside. He has benefitted from the outrageous rule that sees all prisoners serving only half of their sentences (and this is not dependent on “good behaviour”) but still reoffends within a very short time of leaving prison. When considering whether or not “prison works” the first question to be asked is who is it supposed to be working for – the offender or the rest of us? In this case I think “the rest of us” should take precedence.