Statisticians at the Department for Children Schools and Families have been burning the midnight oil to reconcile two sets of crime statistics.

The issue, in simple terms, is whether people under 18 given Penalty Notices for Disorder (PNDs) ought to be counted as First-Time Entrants to the criminal justice system. At present they are not, allowing the Youth Justice Board to claim success in meeting its targets for cutting youth crime.
 
But the PND figures for under-18s are counted by the Office for Criminal Justice Reform as "offences brought to justice", enabling it, too, to meet its targets. One government department achieves its target by counting something which another government department meets its target by discounting. With me so far?
 
There were almost 20,000 PNDs issued to 16 and 17 year-olds in 2007, for offences such as causing harassment, alarm or distress, destroying or damaging property  (so long as the damage is less than £500), and shoplifting (so long as it’s less than £200-worth). Since then possession of a class B drug has been added to the list.
 
The counting/not counting anomaly was pointed out at the beginning of the year, and commented upon here. On July 31, DCSF issued a nine-page Statistical Notice indicating what it intended to do to resolve it. (I missed it at the time, so apologies for the long delay in commenting.)
 
The notice admits there’s a problem. “We appreciate that the difference in the coverage of offences between these two measures could be difficult for the reader to understand”. Actually it’s easy to understand. Accepting it is the difficult bit. 
 
So what’s to be done? PNDs can’t simply be added to the first-time entrant figures, because it could cause double-counting. Lots of juveniles who get PNDs go on to get reprimands, warnings, or convictions, which are counted. A new dataset would be needed.
 
That could be done, to produce a list of first-time entrants that included those who got PNDs. But let’s not rush.
 
“Before committing to this second measure, a good deal of work is needed to establish data quality, presentation and usefulness” – as well as consulting users, in line with the Code of Practice for Official Statistics. (To use the code as a delaying tactic is a Whitehall masterstroke, since it is the guardian of the Code – the UK Statistics Authority – which has pressed for the anomaly to be corrected.)
 
Even if a new dataset is constructed, the targets will remain based on the current measure, so that’s all right. Nobody need worry that the outbreak of honesty will cause them to miss their targets. This second measure would simply be to “aid public understanding of the system”.
 
So it’ll be a while before there is any resolution. And it won’t really change anything. Is that unduly cynical? I think not.